1 INFO-VAX	Wed, 21 Mar 2001	Volume 2001 : Issue 160       Contents: BASIC side effect  Re: BASIC side effect  Re: BASIC side effect P Re: Boring the pants off the COV (was RE: Dumbing Down VMS with UNIX   Elements?P Re: Boring the pants off the COV (was RE: Dumbing Down VMS with UNIXElements? (wC Re: Boring the pants off the COV (was RE: Dumbing Down VMS with...) A Re: Can a VAXStation 4000/96 Boot At Power-up W/O Graphics Card ? D Can anyone point me to a starting point to get some ASE infromation?H Re: Can anyone point me to a starting point to get some ASE infromation?F re: Can VMS "root" from a device other than the one it is booted from?F Re: Can VMS "root" from a device other than the one it is booted from?$ Re: Catalyst for growth? ERP systems$ Re: Catalyst for growth? ERP systems$ Re: Catalyst for growth? ERP systems COE implementation detailsP Re: Compaq ASE Program (was: Re: Can anyone point me to a starting point to get  Re: CSWS and log filter  Re: CSWS and log filter  Re: CSWS and log filter  Re: CSWS and log filter 1 Re: DECnet a proprietary protocol ?  Not at all ! 1 Re: DECnet a proprietary protocol ?  Not at all ! 1 Re: DECnet a proprietary protocol ?  Not at all ! 1 Re: DECnet a proprietary protocol ?  Not at all ! 1 Re: DECnet a proprietary protocol ?  Not at all ! 1 Re: DECnet a proprietary protocol ?  Not at all ! 1 Re: DECnet a proprietary protocol ?  Not at all ! 1 RE: DECnet a proprietary protocol ?  Not at all ! 1 Re: DECnet a proprietary protocol ?  Not at all ! 1 Re: DECnet a proprietary protocol ?  Not at all ! 1 Re: DECnet a proprietary protocol ?  Not at all ! 1 Re: DECnet a proprietary protocol ?  Not at all ! 1 Re: DECnet a proprietary protocol ?  Not at all ! 0 Re: DECnet a proprietary protocol ? Not at all !0 Re: DECnet a proprietary protocol ? Not at all !0 Re: DECnet a proprietary protocol ? Not at all !0 Re: DECnet a proprietary protocol ? Not at all !D Dumbing down comp.os.vms with "Dumbing Down VMS with UNIX Elements?"P Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS       EducationalProgrP Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS       EducationalProgrP Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS       EducationalProgrP Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS      EducationalPrograP Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS      EducationalPrograP Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS      EducationalPrograP Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS      EducationalPrograP Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS     EducationalProgramP Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS     EducationalProgramP Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS     EducationalProgramP Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS     EducationalProgramP Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS     EducationalProgramP Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS     EducationalProgramP Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS     EducationalProgram? Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS  ...) ? Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS  ...) > Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS ...)> Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS ...)P Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS EducationalProgram) PrP Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS EducationalProgram) PrP Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS EducationalProgram) PrP RE: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS EducationalProgram) PrP RE: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS EducationalProgram) PrP RE: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS EducationalProgram) PrP RE: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS EducationalProgram) PrP Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMSEducationalProgram) ProP Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMSEducationalProgram) ProP Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMSEducationalProgram) ProP Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMSEducationalProgram) ProP RE: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMSEducationalProgram) ProP Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMSEducationalProgram) ProP Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMSEducationalProgram) Pro) Re: ElSA Gloria Synergy-8 patch confusion ) Re: ELSA Gloria Synergy-8 patch confusion  Re: Free The OpenVMS Seven! > Help ................... Need Sybase DBA with VMS skills    !! HELP on License Requestor  Info on low level driver access  Re: Internet Printing Protocol Re: Internet Printing Protocol Is there extra logging in BIND? ' Re: login failure when sysuaf is locked ' Re: login failure when sysuaf is locked  Re: MOUNT/BIND too slow  NAS/SAN  Re: OpenVMS Educational Program # Re: POSIX Streams, File Permissions # Re: POSIX Streams, File Permissions # Re: POSIX Streams, File Permissions # Re: POSIX Streams, File Permissions M Re: POSIX Streams, File Permissions (was: Re: The Chris and Bill show. Act II M Re: POSIX Streams, File Permissions (was: Re: The Chris and Bill show. Act II M Re: POSIX Streams, File Permissions (was: Re: The Chris and Bill show. Act II  Re: RMS and file types.  Re: RMS and file types.  Re: RMS and file types.  Re: RMS and file types. . Re: Senior OpenVMS systems manager - SINGAPORE7 Re: STARLETSD and SDL/NOPARSE (was: RE: sytem logicals)  Re: Support of old systems Re: Support of old systems Re: Support of old systems Re: Support of old systems Re: Support of old systems Re: Support of old systems Re: Support of old systems Re: sytem logicals RE: sytem logicals8 The art of flaming (was Re: OpenVMS Educational Program) Re: TURBOchannel graphics  Re: TURBOchannel graphics  Re: uVax3100 m10 & VMS5.2 P VERB SHOW (was Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS Educati, [Announce] Perth WA LUG reformation meeting.6 [Change subject] What if VMS and MacOS had prevailed ? Re: [DCL] minute of the day  Re: [DCL] minute of the day  Re: [DCL] minute of the day  Re: [DCL] minute of the day * Re: [Q] TCP/IP RPC connection error statusC [Random non-DCL VMS quesion of the day] Need a clue about CLUE help G Re: [Random non-DCL VMS quesion of the day] Need a clue about CLUE help   F ----------------------------------------------------------------------  % Date: Wed, 21 Mar 2001 08:57:02 -0700 5 From: "cstranslations" <cstranslations@email.msn.com>  Subject: BASIC side effect) Message-ID: <etFZm9hsAHA.291@cpmsnbbsa09>   K I was working on some code written in BASIC and decided to give something a L try (not expecting the compiler to acctually accept it).  Interesting enoughK BASIC didn't complain. Below is a small sample where the compiler generates L code matching a C style side effect (OpenVMS 7.1-1H2 and 7.2-1, Alpha, BASIC" 1.3 and 1.4). I'm still surprised.  # Just thought I'd pass this along...    Joe    option type = explicit   declare string   SearchStr declare integer  result   
 set no prompt    input "_string: "; SearchStr  ( result = (seg$(SearchStr, 1, 3) = "ABC") if (result) then&    print "The 1st 3 letters are 'ABC'" else<    print "The string does not start with (upper case) 'ABC'" end if print    ------------------------------  % Date: Wed, 21 Mar 2001 09:35:02 -0800 1 From: "Randy Park" <rjpark@mindspring.com.nospam>  Subject: Re: BASIC side effect2 Message-ID: <99aoou$e54$1@slb1.atl.mindspring.net>  0 I've got more years with DEC Basic, Vax Basic, &. Basic Plus II than I can remember.  Personally2 I don't see anything unusual with the code, except that it's in lower case ;-)   - Are you referring to the line beginning with:      result =4 This has always been valid, even in BP2.  Comparison3 operations can be used in an expression. The result 2 of a conditional operation is zero if false and -1% if true.  In this case the comparison #     (seg$(SearchStr, 1, 3) = "ABC") 6 is performed and a zero or -1 is assigned to "result".0 "Result" is like a boolean value.  You then test3 it in your "if" clause and perform your conditional 7 clause.  In the "if" clause, zero if false and non-zero  is true.   Welcome to Basic!!    > cstranslations <cstranslations@email.msn.com> wrote in message# news:etFZm9hsAHA.291@cpmsnbbsa09... K > I was working on some code written in BASIC and decided to give something  a G > try (not expecting the compiler to acctually accept it).  Interesting  enoughC > BASIC didn't complain. Below is a small sample where the compiler 	 generates H > code matching a C style side effect (OpenVMS 7.1-1H2 and 7.2-1, Alpha, BASIC $ > 1.3 and 1.4). I'm still surprised. > % > Just thought I'd pass this along...  >  > Joe  >  > option type = explicit >  > declare string   SearchStr > declare integer  result  >  > set no prompt  >  > input "_string: "; SearchStr > * > result = (seg$(SearchStr, 1, 3) = "ABC") > if (result) then( >    print "The 1st 3 letters are 'ABC'" > else> >    print "The string does not start with (upper case) 'ABC'" > end if > print  >  >    ------------------------------  % Date: Wed, 21 Mar 2001 10:56:00 -0700 5 From: "cstranslations" <cstranslations@email.msn.com>  Subject: Re: BASIC side effect) Message-ID: <OdzBFAjsAHA.354@cpmsnbbsa07>   < "Randy Park" <rjpark@mindspring.com.nospam> wrote in message, news:99aoou$e54$1@slb1.atl.mindspring.net...2 > I've got more years with DEC Basic, Vax Basic, &0 > Basic Plus II than I can remember.  Personally4 > I don't see anything unusual with the code, except > that it's in lower case ;-)   K That does seem to be the case with all of the BASIC code I've ever loked at  unless I wrote it :-)    > / > Are you referring to the line beginning with:  >     result =6 > This has always been valid, even in BP2.  Comparison5 > operations can be used in an expression. The result 4 > of a conditional operation is zero if false and -1	 > if true   F I know (knew) it would work after a statement with an "attached" IF or UNLESS. For example    PRINT "TRUE" IF (1 = 1)   H though I suspect the compiler would do a bit of optimizing on the above.   > Welcome to Basic!!  J Well - I've been working in *primarily* BASIC shops for almost 10 years atG this point (longer than I care to remember). Most of the code that I've K supported (and currently support) was written when I was in the 7th and 8th 	 grade :-)   @ Seen some strange things. Never ran into the below in any of it.  L If they would add a pointer/derefence mechanism that would be rather useful.  But then that'll never happen...   Joe    > @ > cstranslations <cstranslations@email.msn.com> wrote in message% > news:etFZm9hsAHA.291@cpmsnbbsa09... C > > I was working on some code written in BASIC and decided to give 	 something  > a I > > try (not expecting the compiler to acctually accept it).  Interesting  > enoughE > > BASIC didn't complain. Below is a small sample where the compiler  > generates J > > code matching a C style side effect (OpenVMS 7.1-1H2 and 7.2-1, Alpha, > BASIC & > > 1.3 and 1.4). I'm still surprised. > > ' > > Just thought I'd pass this along...  > >  > > Joe  > >  > > option type = explicit > >  > > declare string   SearchStr > > declare integer  result  > >  > > set no prompt  > >   > > input "_string: "; SearchStr > > , > > result = (seg$(SearchStr, 1, 3) = "ABC") > > if (result) then* > >    print "The 1st 3 letters are 'ABC'" > > else@ > >    print "The string does not start with (upper case) 'ABC'"
 > > end if	 > > print  > >  > >  >  >    ------------------------------  % Date: Wed, 21 Mar 2001 15:38:19 +0000 % From: Alan Greig <a.greig@virgin.net> Y Subject: Re: Boring the pants off the COV (was RE: Dumbing Down VMS with UNIX   Elements? 8 Message-ID: <c8ihbt4c1glsr0uu9teefjdca4mkc5ehhb@4ax.com>  2 On Wed, 21 Mar 2001 10:20:20 +0000, Christof Brass <brass@infopuls.com> wrote:     @ >The UNIXification of VMS already started. That's why we have to( >express our concerns and fight against.  E No we don't as nothing will be done to break backwards compatibility. A Where new changes are required to the operation of underlying VMS E features to support COE you will be able to turn these features on or D off. For example $ SET VOLUME/ENABLE=ACCESS_DATES is required beforeB VMS starts retaining read dates.  $ SET VOLUME/ENABLE=HARDLINKS toE enable unix style hard links  $ SET VOLUME/CASE_LOOKUP=SENSITIVE etc.   - This from the new features manual for 7.2-6C1    >  -- Alan   ------------------------------  % Date: Wed, 21 Mar 2001 10:20:20 +0000 ) From: Christof Brass <brass@infopuls.com> Y Subject: Re: Boring the pants off the COV (was RE: Dumbing Down VMS with UNIXElements? (w , Message-ID: <3AB88064.EF5F3318@infopuls.com>   "Doc.Cypher" wrote:  > $ > -----BEGIN PGP SIGNED MESSAGE----- > C > On Tue, 20 Mar 2001, Christopher Smith <csmith@amdocs.com> wrote:  > >> -----Original Message----- = > >> From: bdwheele@indiana.edu [mailto:bdwheele@indiana.edu]  > > H > >[I will only comment on the beginning of this incredibly long message
 > >-Chris] > 2 > I won't, I'll comment on the entire damn thread. > B > >> care to point to these analysis?  I submit to you that 80% of > >> all statistics  > >> on usenet are made up.  > >  > >Yep.  Or more.  >  > Actually, no . . . > 3 >         According to the latest official figures, 8 >         43.6% of all statistics are totally worthless. >  > >> <sarcasm>< > >> Oh, and nobody has ever written a bad app for VMS, they > >> always stick to > >> unix for that!  > >> </sarcasm>  > > N > >Well, it could be argued the VMS's design makes it more difficult (assumingD > >you're after a _functional_ app) to write poor code in some ways. > K > I've been following this "VMS is the best thing since Alan Turing's time"  > argument for too long now.   Misunderstanding.   M > If COE is the lifeline that might help bring VMS out of relative obscurity, J > fine. End of Discussion. Please! [Even Andrew seems to agree with that!]  ) With this method it wouldn't VMS anymore.   M > Does it mean VMS will become UN*Xified? No, never. It will remain different M > from UN*X (that is, not better in all circumstances). this should be simple I > and obvious. Security through obscurity might suit some, but it doesn't  > wash for serious usage.   ? The UNIXification of VMS already started. That's why we have to ' express our concerns and fight against.   C > VMS has *no* job security if it remains hidden in obscure places.  > I > Please, please, remember that it isn't about survival of the fittest in G > terms of elegance or robustness. Numerical superiority has a habit of 3 > winning when you are fighting a war of attrition.    We don't need another UNIX.   J > Anyway, the options and issues discussed in this thread simply look likeL > nonsense. I've still got my sharp engineer's pencil - and that solves moreH > problems than *any* sort of computer. We are discussing tools, and theM > black vs. white arguments promoted herein look like comparing square wheels M > with triangular ones. They're all round, but some come with pneumatic tyres 7 > (unlike Weeniedoze which doesn't survive bumpy apps).   < We need different tools. We don't need to change one tool to@ another. From your abstract point of view this should make sense to you.    > Doc. >  > -----BEGIN PGP SIGNATURE-----  > Version: 2.6.2 > B > iQEVAwUBOraPcsriC3SGiziTAQH2ZQf+Jr8mwZLdtmpkNG9Go6mvnzbFj6A4ipL+B > 8EbvRsW8QmLrSnHjP4e3gw3/GMJ+XWnzRIwIkcXTc6k1pttCKR23qz86ZO6GHajlB > fPxW/1q2vWUJbUSwI7aq5qZblDoPqcRoN6QiPwoXR/S+1guyUpJMbhGh0mCXNGH5B > 5igPLLGBwkpBMpo5CTp14iOmDMpYn0D6PWo932hLlXXAhANd7gGYhlv8Xxu+/wGzB > Wocb7GNqvNkxd8SVgXGMhuqtHeiWw0Fr16X1S/EmV/vnBggOvO8ruybks3xyAvbJ: > kme9OQDr9oFjtUjBVLcqxy6gTbUDsaUvPiXxKYkvGjmYFLL6o6c+lA== > =oSeb  > -----END PGP SIGNATURE-----    ------------------------------    Date: 21 Mar 2001 12:22:41 -00004 From: Doc.Cypher <Use-Author-Address-Header@[127.1]>L Subject: Re: Boring the pants off the COV (was RE: Dumbing Down VMS with...)0 Message-ID: <WakeUpPlease.Christoff@nowhere.nil>  " -----BEGIN PGP SIGNED MESSAGE-----  ? On Wed, 21 Mar 2001, Christof Brass <brass@infopuls.com> wrote:    <snip>  L >> I've been following this "VMS is the best thing since Alan Turing's time" >> argument for too long now.e >t >Misunderstanding.  F Really? You do seem to be in favour of VMS in an almost fanatical way.  N >> If COE is the lifeline that might help bring VMS out of relative obscurity,K >> fine. End of Discussion. Please! [Even Andrew seems to agree with that!]t > * >With this method it wouldn't VMS anymore.  J No, it would be VMS plus. Plus a modicum of standardization. Plus a commonK API which UN*X will also have. Essentially, plus something pretty sensible."J It *will* permitting easy porting of applications. Underlying the solution from "The Q" would be VMS.  N >> Does it mean VMS will become UN*Xified? No, never. It will remain differentN >> from UN*X (that is, not better in all circumstances). this should be simpleJ >> and obvious. Security through obscurity might suit some, but it doesn't >> wash for serious usage. >f@ >The UNIXification of VMS already started. That's why we have to( >express our concerns and fight against.  H VMS cannot become "just another UN*X". VMS will not become "just anotherG UN*X". And, I'd agree with you when you say VMS should not become "justaB another UN*X", but you're being rabid about this. Look upon COE asK expecting UN*X to raise their standards (or at least find some) rather thano VMS lowering its standard.  D >> VMS has *no* job security if it remains hidden in obscure places. >>  J >> Please, please, remember that it isn't about survival of the fittest inH >> terms of elegance or robustness. Numerical superiority has a habit of4 >> winning when you are fighting a war of attrition. >  >We don't need another UNIX.  5 You do - you obviously don't like the one you've got.a  K >> Anyway, the options and issues discussed in this thread simply look likeuM >> nonsense. I've still got my sharp engineer's pencil - and that solves more I >> problems than *any* sort of computer. We are discussing tools, and theuN >> black vs. white arguments promoted herein look like comparing square wheelsN >> with triangular ones. They're all round, but some come with pneumatic tyres8 >> (unlike Weeniedoze which doesn't survive bumpy apps). >E= >We need different tools. We don't need to change one tool to A >another. From your abstract point of view this should make senseE >to you.    F COE isn't about tools. It is about toolmaking. Consider it in the sameG manner as universal adoption of the Metric System. You might prefer theNE old, but you're as likely to win fighting progress as King Canute wasV trying to command the tides.     Doc.   -----BEGIN PGP SIGNATURE-----n Version: 2.6.2  @ iQEVAwUBOrfg8sriC3SGiziTAQHn8Qf/c1qnOG403ru61tiJNPofVDFvbWWvZrX5@ rI0QxN7icRK1dl+75YdaSlrXJzoWcjrUjMf7pajqR4h6aJ2a/2eoht/UV3Geor0A@ HUuDAhbOTyV5TkSE4u59N8gwQSxGj2SIAbbphk8CJLi9RZFBRJWwQyy7PhPt/F+r@ drTaNvL5/QJlKfJqUam45b+nl7oxgcXQNRgP5g9VGmNw8EMoonynT5ZI9/SaaFPC@ a/ee1f0jHxPC+UjWe74Ii2PgpGLX37odfC/Y11Muuij7CCpUir+RgDz8EP6qGONJ8 Kok9drooNCQD5a1iejXpKkWDaCXAAGZvI8HAMjYy3sEN+PKeDV4CjA== =Y6gRo -----END PGP SIGNATURE-----a   ------------------------------  # Date: Wed, 21 Mar 2001 07:29:18 GMTp4 From: LESLIE@209-16-45-102.insync.net (Jerry Leslie)J Subject: Re: Can a VAXStation 4000/96 Boot At Power-up W/O Graphics Card ?( Message-ID: <iLYt6.639$%L5.14591@insync>  - Bruce J Baker (bjbaker@bigpond.net.au) wrote:r8 : "Jerry Leslie" <leslie@clio.rice.edu> wrote in message# : news:996vol$hso$1@joe.rice.edu... @ : > The systems is a VAXStation 4000/96 running VMS 7.1, without* : > a graphics device, mouse, or keyboard. : > I : > The "HALT" flag is set to "2", which should cause  the system to boot?H : > on power-up.  But the system halts at the three-chevron prompt with: : >t" : >    ?? 001    2     LCSPX  0512 : >rG : > It can be manually booted at that point. The graphics card has beeno : > reseated several times.S : > : If memory serves, you can get past this with >>> SET FBOOT 1  : which bypasses the boot tests. :   " Thanks, Bruce, that did the trick.+ And thanks to every one else who responded.   / --Jerry Leslie   leslie@209-16-45-97.insync.netn;                  leslie@209-16-45-102.insync.net is invalidP   ------------------------------  % Date: Wed, 21 Mar 2001 08:57:54 -0700g  From: Jon <jsmyth69@hotmail.com>M Subject: Can anyone point me to a starting point to get some ASE infromation?w2 Message-ID: <FM+4OmerAJJ7aHQjnO9dUKbcNJcD@4ax.com>   Hi,   E   I was told at work that I could attempt to get an ASE certificationc? and was wondering where to start looking for information.  I am.+ looking for classes, costs, etc...  Thanks!P   Can you help me out?   o   ------------------------------  % Date: Wed, 21 Mar 2001 12:05:17 -0500P) From: "Bob Lail" <robert.lail@compaq.com> Q Subject: Re: Can anyone point me to a starting point to get some ASE infromation?l. Message-ID: <mb5u6.2$fB6.264@news.cpqcorp.net>   Jon    Start with this url;  + http://www.compaq.com/training/aseinfo.htmlt  	 \Bob Laile        - "Jon" <jsmyth69@hotmail.com> wrote in messageO, news:FM+4OmerAJJ7aHQjnO9dUKbcNJcD@4ax.com... > Hi,l >eG >   I was told at work that I could attempt to get an ASE certificationFA > and was wondering where to start looking for information.  I amI- > looking for classes, costs, etc...  Thanks!  >  > Can you help me out?   ------------------------------  % Date: Wed, 21 Mar 2001 10:48:44 +0000e/ From: Nigel Arnot <sysmgr@maxwell.ph.kcl.ac.uk>sO Subject: re: Can VMS "root" from a device other than the one it is booted from?e7 Message-ID: <009F957C.F700AC40.22@maxwell.ph.kcl.ac.uk>   H > I'm used to UNIXes, sorry...  Anyway, I have a VAXserver 3100 (unknownL > model, but it's unable to boot from a CD-ROM), a few disks for it (210 megI > each), and an OpenVMS hobbyist CD-ROM.  It originally booted a somewhataL > damaged VAX/VMS V5.5 installation, but I broke that...  My idea was to tryL > putting standalone backup from one of the other harddisks (which went OK),J > then copy the savesets from the CD to the standalone backup disk, (whichH > also went OK), boot the disk (which also went OK) and then restore theL > savesets that way (which went really, really badly).  Of course, having noI > backup media, I couldn't back up my disk first, so now I have a machinesK > that can only boot V5.5 standalone backup.  The thing is, UNIXes can boot L > from one disk, and then "root", or use as a system disk, some other device8 > in the system.  Can VMS do that, or am I just screwed?  I VMS needs a VMS-compatible "disk" to boot off. With a 3100 this can be a  D VMS-compatible SCSI disk local to the system, or an MSCP-served disk= within the LAVC cluster that it's booting as a satellite of. g  B The latter probably isn't of interest to you (though if you have aC friendly VMScluster and sysadmin available, it's by far the easiestpK way to go, and probably how VMS got onto a 210Mb disk in the first place). SH The former is a problem with such a small disk, as (I think) it's below K the minimum size that is supported (certainly is for current VMS versions).r  I Your easiest path might be to obtain an RZ26 (1Gb) or RZ28 (2Gb), they'retF pretty cheap these days on the 2nd-user market or you might be able toD scrounge one. Non-DEC disks can be problematic, especially as systemI disks, because if they prove to be incompatible with whatever VMS version-K you want to upgrade to, you are stuck again with an unbootable wreck. Also,PJ they have to be compatible with the bootstrap driver as well as the normal< SCSI driver that gets loaded off the system disk after boot.  M In answer to the other question: once a minimal VMS installation has booted, yA it's possible (though unsupported) to play with the definition oflH SYS$SYSROOT to make it multi-valued. In a cluster node it's 2-valued, toH search first the cluster node root and then the cluster common root. YouA could instead split the system over two or more local roots. It'siM reputed to work, but is unsupported, and therefore getting software installedhK into such an environment is nontrivial. (That's nontrivial, not impossible; K you just have to do some work that Digital/Compaq didn't because they don'tp  support this mode of operation).  I I'd still recommend the RZ28; I've seen them listed by a broker at UK# 75c6 (which probably means you can haggle down from there).  L > In the latter case, how hard would it be to write a new bootstrap program,& > and load it over the serial console? >    Very?    	Yours,t
 		Nigel Arnotc- 		NRA@MAXWELL.PH.KCL.AC.UK                      7 		"In the beginning there was nothing, which exploded."e   ------------------------------  % Date: Wed, 21 Mar 2001 12:01:31 +0000S  From: steven.reece@quintiles.comO Subject: Re: Can VMS "root" from a device other than the one it is booted from? H Message-ID: <OF8518DFA3.A7458220-ON80256A16.0041E7A1@qedi.quintiles.com>  E I think the change was made in either 6.2 or 7.1 (I suspect 7.1 but Ig( haven't a manual to hand at the moment).I At least one of my previous managers swore black was white that it wasn't  the case......   Antonio Carlini commented:- >>>The most recent OpenVMS VAX CDs (from V7.2r- onwards certainly, possibly further back too) . are also bootable ... try booting root [SYS1]. <<<a   ------------------------------  % Date: Wed, 21 Mar 2001 10:03:09 +0000 ) From: Christof Brass <brass@infopuls.com> - Subject: Re: Catalyst for growth? ERP systemsr, Message-ID: <3AB87C5D.B8C41A1B@infopuls.com>   Alan Greig wrote:e > C > On 20 Mar 2001 14:33:18 GMT, bdwheele@indiana.edu (Brian Wheeler)r > wrote: > Q > >YOU are missing the point:  VMS cannot continue to do its own thing and expectcO > >to magically come out of the dark and be everyone's favorite OS.  A catalystkK > >is required.  This catalyst may very well be the COE initiative.  It mayn3 > >not.  But not trying it isn't going to help any.d > H > I see COE as a good idea but the major catalyst to stimulate VMS salesF > would be a port of SAP R3. SAP running on top of RDB would blow awayD > the competition if done properly. And just maybe the COE work will" > make a SAP port that bit easier, > C > Worldwide we use two major MRP/ERP systems. MANMAN and SAP. A SAPiG > implementation of ours in Norway running on Wintel (EDS specified and C > manage the system) does not perform up to expectations and we area? > trying to suggest Alphaservers. Trouble is that the corporatedE > standards say that HP-UX is the corporate Unix standard... It would-H > actually be easier for us to argue for SAP on VMS (VMS still being theA > platform for most MANMAN sites) than it is to argue for Tru-64.r > @ > With Oracle Apps about to disappear from the VMS portfolio andF > PeopleSoft (I believe) already gone the availability of state of theH > art ERP systems continues to decline towards vanishing point. There isD > no more natural home for these type of applications than VMS in my
 > opinion. > D > Despite pressing Compaq for a port I would have placed this in theH > highly unlikely category until recently but with ads for VMS appearing0 > in the UK press last week maybe pigs will fly! >  > -- > Alan  @ You are so right - I and several others mentioned the importance< of having SAP on VMS. But there is one clue which would make= this even more attracting. SAP suffers from not having a realr> architectures solution for Web access. The current solution is8 completely crap and a waste of computing power and human
 resources.? SAP could do two things together - instead of merely porting toe; VMS: they could implement a proper Web access layer on VMS.A  @ Do you know that the SAP founders are former IBM people (I think they hated DEC)?? Do you know that the SAP people now have very tight connections@@ to M$? And vice versa as Bill Gates publicly stated last year or the year before.   ------------------------------  % Date: Wed, 21 Mar 2001 09:53:15 +0000o% From: Alan Greig <a.greig@virgin.net>0- Subject: Re: Catalyst for growth? ERP systems 8 Message-ID: <rksgbtg9povst3899s9rs62f1o4h94jl6q@4ax.com>  F On 20 Mar 2001 17:15:44 -0500, koehler@encompasserve.org (Bob Koehler) wrote:  a >In article <7bsebt4bj3ppeevp68rforlc99h4taqpqk@4ax.com>, Alan Greig <a.greig@virgin.net> writes:r >>  I >> I see COE as a good idea but the major catalyst to stimulate VMS sales G >> would be a port of SAP R3. SAP running on top of RDB would blow awayME >> the competition if done properly. And just maybe the COE work wille# >> make a SAP port that bit easier,C >_H >Somehow I don't think RDB's API is one that's likely to show up in COE.  D No but I wouldn't expect the use of calls to the COE API to precludeE calls to RDB from the same source code any more than I would expect C:D library calls to preclude calls to Oracle. As RDB supports SQL/Net IF would hope that not much code changes would be required in SAP to talk= to RDB as opposed to native Oracle. For example we use OracleTF Developer 2000 applications on VMS to talk directly to an RDB databaseF (part of MANMAN backend database). Developer 2000 thinks it is talking to a native Oracle database.  D What I would hope is that you could take SAP for Tru-64 (or Linux orC Solaris) and just build it without too much difficulty under VMS. I F have been told that it is a goal of the COE project to allow just thisF sort of thing. If it didn't then the whole project is close to a waste of time in my opinion.  C Such a port as described above would allow SAP to run under COE but3? the underlying database (where all the real work is done) to be  native. I believe RDB 8 (orwB 7.whatever-they-have-decided-to-name-it-now) enhances even further* RDB's ability to look like native Oracle.   G >----------------------------------------------------------------------r@ >Bob Koehler                     | Computer Sciences Corporation> >NASA GSFC Flight Software       | Federal Sector, Civil GroupF >                                | please remove ".aspm" when replying   -- Alan   ------------------------------  % Date: Wed, 21 Mar 2001 15:28:26 +0000 % From: Alan Greig <a.greig@virgin.net>m- Subject: Re: Catalyst for growth? ERP systemso8 Message-ID: <oghhbtcl3raicp6l6d01n4609ggd1j8emt@4ax.com>  2 On Wed, 21 Mar 2001 10:03:09 +0000, Christof Brass <brass@infopuls.com> wrote:l   >cA >You are so right - I and several others mentioned the importance = >of having SAP on VMS. But there is one clue which would makeu> >this even more attracting. SAP suffers from not having a real? >architectures solution for Web access. The current solution is 9 >completely crap and a waste of computing power and human  >resources.   F Are you talking about mysap (the relatively new web access screens) orC the older halfway solutions via weblinks to custom written code? IfiD mysap then it may be a waste of computing power but it is selling by, the bucket-load (just like crap I suppose).   @ >SAP could do two things together - instead of merely porting to< >VMS: they could implement a proper Web access layer on VMS. >tA >Do you know that the SAP founders are former IBM people (I thinkc >they hated DEC)?t@ >Do you know that the SAP people now have very tight connectionsA >to M$? And vice versa as Bill Gates publicly stated last year or   D It is true that Bill Gates stated this and I asked about this at SAPA UK a couple of months ago. SAP still only recommend Microsoft SQLr@ Server in low end configurations despite what Bill might have us believe-   >the year before.E   -- Alan   ------------------------------    Date: 22 Mar 2001 01:58:58 +0800, From: Paul Repacholi <prep@prep.synonet.com># Subject: COE implementation details 0 Message-ID: <87k85j80m5.fsf_-_@prep.synonet.com>  ' Alan Greig <a.greig@virgin.net> writes:m  ? > $ SET VOLUME/ENABLE=HARDLINKS to enable unix style hard linkst  E But VMS already has 'hard links'. No ref counts, but it does have thea6 links.  So what is the change if you enable hardlinks?   -- -< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.@@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  # Date: Wed, 21 Mar 2001 18:42:56 GMT12 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)Y Subject: Re: Compaq ASE Program (was: Re: Can anyone point me to a starting point to get "/ Message-ID: <QC6u6.13$fB6.354@news.cpqcorp.net>a  U In article <FM+4OmerAJJ7aHQjnO9dUKbcNJcD@4ax.com>, Jon <jsmyth69@hotmail.com> writes:   F :  I was told at work that I could attempt to get an ASE certification@ :and was wondering where to start looking for information.  I am, :looking for classes, costs, etc...  Thanks!  <   "The Compaq Accredited Systems Engineer (ASE) Program is a@   worldwide, Compaq- specific certification program recognizing *   technical proficiency in integrating..."  
   Details at:e       http://www.compaq.com/ase/    N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------    Date: 21 Mar 2001 13:03:29 +0100G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>.  Subject: Re: CSWS and log filterH Message-ID: <y48zlz9vn2.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  > "Martin Vorlaender" <martin.vorlaender@pdv-systeme.de> writes:  6 > piped_log_spawn: unable to exec /bin/sh -c 'cronologC >    /apache$specific/logs/AP%y%m%d.log': no such file or directory  [snip]" > Directory APACHE$COMMON:[000000] > G > CRONOLOG.EXE;1       [WEB,APACHE$WWW]                 (RWED,RWED,RE,)h  B Make sure the exec above know about cronolog - it has no directoryG specification, so are you relying on the current default directoy being ; APACHE$COMMON:[000000] or that directory being in DCL$PATH?t   	Jan   ------------------------------  % Date: Wed, 21 Mar 2001 07:29:30 -0500h3 From: "Gaitan D'Antoni" <gaitan.dantoni@compaq.com>y  Subject: Re: CSWS and log filter1 Message-ID: <791u6.189$eE2.9968@news.cpqcorp.net>u  K It appears that support for the pipe feature of the CustomLog directive dideJ not make it into the OpenVMS version of the Apache server. We'll look into( adding this feature in a future release.  I CSWS V1.1, due to be released for beta testing in a couple of weeks, doeseL include some log file enhancements that will allow you to close existing logI files and open new ones in a running server. This way you will be able to.D examine or archive log files without having to shut down the server.   -- Gaitan D'AntoniS. Apache Web Server for OpenVMS Technical LeaderC http://www.openvms.compaq.com/openvms/products/ips/apache/csws.htmln Compaq Computer Corporation-   ------------------------------  % Date: Wed, 21 Mar 2001 18:08:21 +0100Q< From: "Martin Vorlaender" <martin.vorlaender@pdv-systeme.de>  Subject: Re: CSWS and log filter3 Message-ID: <99an6j$fiku$1@ID-56200.news.dfncis.de>B   Jan Vorbrueggen wrote...? >"Martin Vorlaender" <martin.vorlaender@pdv-systeme.de> writes:x7 >> piped_log_spawn: unable to exec /bin/sh -c 'cronologoD >>    /apache$specific/logs/AP%y%m%d.log': no such file or directory >[snip]G# >> Directory APACHE$COMMON:[000000]- >> -H >> CRONOLOG.EXE;1       [WEB,APACHE$WWW]                 (RWED,RWED,RE,) >.C >Make sure the exec above know about cronolog - it has no directoryeH >specification, so are you relying on the current default directoy being< >APACHE$COMMON:[000000] or that directory being in DCL$PATH?    < I tried both: adding a path to the cronolog pipe, and adding' APACHE$COMMON:[000000] to the DCL$PATH.o  F I'll have to wait for CSWS 1.1 to not shutdown the web server in order. to rename the daily logs (see Gaitan's reply).   cu,o   Martin -- sJ One OS to rule them all       | Martin Vorlaender  |  VMS & WNT programmer7 One OS to find them           | work: mv@pdv-systeme.de J One OS to bring them all      |   http://www.pdv-systeme.de/users/martinv/? And in the Darkness bind them.| home: martin@radiogaga.harz.de e   ------------------------------  % Date: Wed, 21 Mar 2001 18:10:08 +0100h< From: "Martin Vorlaender" <martin.vorlaender@pdv-systeme.de>  Subject: Re: CSWS and log filter3 Message-ID: <99an9v$fft3$1@ID-56200.news.dfncis.de>l   Gaitan D'Antoni wrote...J >CSWS V1.1, due to be released for beta testing in a couple of weeks, doesI >include some log file enhancements that will allow you to close existing  log J >files and open new ones in a running server. This way you will be able toE >examine or archive log files without having to shut down the server.e    H Thanks for that - it'll certainly be easier than the *ix mumbo jumbo ;-) Need another beta tester?n   cu,l   Martin --J One OS to rule them all       | Martin Vorlaender  |  VMS & WNT programmer7 One OS to find them           | work: mv@pdv-systeme.deLJ One OS to bring them all      |   http://www.pdv-systeme.de/users/martinv/> And in the Darkness bind them.| home: martin@radiogaga.harz.de   ------------------------------  % Date: Wed, 21 Mar 2001 11:40:16 +0100.+ From: "Robert de Boer" <r.e.deboer@kpn.com> : Subject: Re: DECnet a proprietary protocol ?  Not at all !. Message-ID: <99a0pc$a4i$1@info.service.rug.nl>   Roy,  J An interesting fact that "several third-party implementations of DECnet onH other platforms" have exist or even are existing, but are you sure about
 that fact?  I I have worked with DECnet-stack implementations for MS-DOS, but that were-K implementations from Digital to support Digitals Pathworks-architecture andlJ in fact proprietary implementations. Perhaps Digtal has made for Macintosh= also DECnet-implementations to support Pathworks-arhitecture.l  G I also know that there are OSI-stack implementations availbale on othervF platforms such als Tandem, Unisys, SUN, OpenVMS (DECnet/OSI from earlyA 1990's), etcetera, but this are not DECnet-stack implementations.t   Could you be mistaken?  
 with regards,A Robert de Boer r.e.deboer@kpn.com, "Roy Omond" <Roy@Omond.net> wrote in message# news:3AB7920F.F405FE74@Omond.net...d > Bill Gunshannon wrote: >n/ > [... huge snip of the usual boring stuff ...]b >uI > > ??  What technical examples??  Name some specific "missing services",nF > > "missing concepts" and uncomplied with "standards".  But remember,H > > not implementing DECNET is not a Unix idea.  It is a proprietary DECF > > protocol and even without any assitance from the only one with the- > > actual definition it is still being done.d >e@ > You hit one of my pet peeves, and my peeves didn't like it :-) >.@ > DECnet is not and never has been a "proprietary DEC protocol".A > It is the very definition of the complete opposite.  The DECneteA > protocol definitions have *always* been free for anyone to makeLD > whatever use of them they desired.  There were several third-partyD > implementations of DECnet on other platforms (I recall at least on > Sun and Macintosh).e >t* > Shame on you for spreading yet-more-FUD. >  > Roy Omondn > Blue Bubble Ltd.   ------------------------------    Date: 21 Mar 2001 20:17:38 +0800, From: Paul Repacholi <prep@prep.synonet.com>: Subject: Re: DECnet a proprietary protocol ?  Not at all !- Message-ID: <87pufbb9jx.fsf@prep.synonet.com>)  - "Robert de Boer" <r.e.deboer@kpn.com> writes:     B > An interesting fact that "several third-party implementations ofE > DECnet on other platforms" have exist or even are existing, but areo > you sure about that fact?a  F > I have worked with DECnet-stack implementations for MS-DOS, but that7 > were implementations from Digital to support Digitalsh0 > Pathworks-architecture and in fact proprietary= > implementations. Perhaps Digtal has made for Macintosh alsoe: > DECnet-implementations to support Pathworks-arhitecture.  E I instaled the first DECnet-DOS in .au. IT was DECnet only. PathworksnA did not exist then. I have a Sun Decnet in the next room, for old-H machines. Up until 199<something>, NASA, the Accelerator Labs ( HEPNET )3 and many large companies run DECnet as THE network.r  K Ki Research (sp?) did DECnet IV for nearly everything that opened and shut.t   -- f< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.e@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------    Date: 21 Mar 2001 08:39:48 -0500- From: koehler@encompasserve.org (Bob Koehler)e: Subject: Re: DECnet a proprietary protocol ?  Not at all !3 Message-ID: <JzMSKCNIifFN@eisner.encompasserve.org>   \ In article <99a0pc$a4i$1@info.service.rug.nl>, "Robert de Boer" <r.e.deboer@kpn.com> writes: > Roy, > L > An interesting fact that "several third-party implementations of DECnet onJ > other platforms" have exist or even are existing, but are you sure about > that fact?   I am.  m  E We are using KiResearch's KiNet right now on our AIX systems and they.H ship KiNet for just about every platform I could conceive of buying.  HPG offered to sell me a DECnet stack from CDC for HP-UX.  Sun has had it'sdE own DECnet for Solaris for years.  We bought and used DECnet productswG for our IBM mainframes from Interlink.  You can crank up DECnet on moste Linux systems.  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporation-= NASA GSFC Flight Software       | Federal Sector, Civil Group E                                 | please remove ".aspm" when replyingr   ------------------------------   Date: 21 Mar 2001 14:09:42 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)M: Subject: Re: DECnet a proprietary protocol ?  Not at all !, Message-ID: <99acn6$2oh6$1@info.cs.uofs.edu>  ) In article <3AB7920F.F405FE74@Omond.net>,a"  Roy Omond <Roy@Omond.net> writes: |> Bill Gunshannon wrote:l |>  0 |> [... huge snip of the usual boring stuff ...] |>  J |> > ??  What technical examples??  Name some specific "missing services",G |> > "missing concepts" and uncomplied with "standards".  But remember,lI |> > not implementing DECNET is not a Unix idea.  It is a proprietary DEC G |> > protocol and even without any assitance from the only one with thea. |> > actual definition it is still being done. |> eA |> You hit one of my pet peeves, and my peeves didn't like it :-)n |> uA |> DECnet is not and never has been a "proprietary DEC protocol".s  = I'll admit I could have been wrong.  I just remembered seeing A somewhere that the reason it took so long to get even rudimentaryaB DECNET into Linux was becuase DECNET was un-published, proprietaryC DEC property and had to be "clean room" reverse engineered.  I knowtC that I have never seen any DECNET code or specs published anywhere.i  B |> It is the very definition of the complete opposite.  The DECnetB |> protocol definitions have *always* been free for anyone to make' |> whatever use of them they desired.  s  C But. as I said, i have never sene them or seen any mention of them.p@ IP on the other hand is published in a dozen textbooks and other places as well.u  E |>                                     There were several third-party E |> implementations of DECnet on other platforms (I recall at least on  |> Sun and Macintosh).  C I remmember the Sun one, but I thought that was a licensed product.hB It defeinitely was not a prat of the base OS on nay version I have
 ever used.   |> P+ |> Shame on you for spreading yet-more-FUD.e   If it is, I apologize.   bill   -- lJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   r   ------------------------------   Date: 21 Mar 2001 14:52:36 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)o: Subject: Re: DECnet a proprietary protocol ?  Not at all !, Message-ID: <99af7k$2pl0$1@info.cs.uofs.edu>  - In article <87pufbb9jx.fsf@prep.synonet.com>,o/  Paul Repacholi <prep@prep.synonet.com> writes:: |> CN |> Ki Research (sp?) did DECnet IV for nearly everything that opened and shut.  E A quick search of the web showed numerous comments that the Spec for oC DECNET Phase IV is publicly available.  I see no mention in any doc@B found about previous versions.  Is it possible that DEC was not asD generous with previous versions??  There is no doubt that Unix boxesG can talk DECNET, it has been a part of Ultrix since the Ultrix-11 days.tH Given this, if it were in fact freely avaialble, why would it have takenG more than a decade for the first non-commercial DECNET for unix to shown- up and that still in a very incomplete form??w   bill   -- tJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------    Date: 21 Mar 2001 16:08:14 +0100G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>a: Subject: Re: DECnet a proprietary protocol ?  Not at all !H Message-ID: <y4k85j88ip.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  3 bill@triangle.cs.uofs.edu (Bill Gunshannon) writes:)  ? > I'll admit I could have been wrong.  I just remembered seeingbC > somewhere that the reason it took so long to get even rudimentarytD > DECNET into Linux was becuase DECNET was un-published, proprietaryE > DEC property and had to be "clean room" reverse engineered.  I knowtE > that I have never seen any DECNET code or specs published anywhere.i  E All of DECnet-IV's specifications were easily available from DEC at adG very reasonable price - so reasonable that the place I used to work for M (a university institute running a 780 to service ~100 users intially) bought -$ them without a really pressing need.  G > |>                                     There were several third-partylG > |> implementations of DECnet on other platforms (I recall at least ona > |> Sun and Macintosh). > E > I remmember the Sun one, but I thought that was a licensed product.0  ; Sure, somebody did the work, so they wanted money for it...l  N There is the amusing story that the DECnet specification had ambiguous wordingJ in what certain fields in network packets would contain in certain (mostlyL error) conditions. One day, a non-DEC PC-based implementation decided to putJ -1 where the VAX implementation put 0. That lead to an access violation inM kernel mode on the receiving end of that packet...they programmed the networkl, stuff a bit more defensively from then on...   	Jan   ------------------------------  # Date: Wed, 21 Mar 2001 14:08:09 GMT / From: "Richard L. Dyson" <rick-dyson@uiowa.edu>c: Subject: Re: DECnet a proprietary protocol ?  Not at all !) Message-ID: <3AB86F79.249FA43B@uiowa.edu>a   Bob Koehler wrote: > ^ > In article <99a0pc$a4i$1@info.service.rug.nl>, "Robert de Boer" <r.e.deboer@kpn.com> writes: > > Roy, > > N > > An interesting fact that "several third-party implementations of DECnet onL > > other platforms" have exist or even are existing, but are you sure about > > that fact?  < 	It was even available for the Amiga platform at one time...   Rick -- tH Richard L. Dyson                                    rick-dyson@uiowa.eduH  _   _      _____                http://www-pi.physics.uiowa.edu/~dyson/H | | | |    |_   _|   Systems Analyst                     O: 319/335-1879H | | | | of   | |     The University of Iowa            FAX: 319/335-17536 | \_/ |     _| |_    Department of Physics & Astronomy-  \___/     |_____|   Iowa City, IA 52242-1479p   ------------------------------  % Date: Wed, 21 Mar 2001 09:36:46 -0600g+ From: Christopher Smith <csmith@amdocs.com> : Subject: RE: DECnet a proprietary protocol ?  Not at all !L Message-ID: <3B55D7F383B0D31197D9009027541CBF0BDD54B7@cmiexch1.cmi.itds.com>   > -----Original Message-----2 > From: Robert de Boer [mailto:r.e.deboer@kpn.com]  @ > An interesting fact that "several third-party implementations  > of DECnet on@ > other platforms" have exist or even are existing, but are you  > sure about > that fact?  F There is at least one independent implementation for linux.  One couldD assume that if they can get the documentation, so can somebody else.   Regards,   Chrisf  ! Christopher Smith, Perl Developera Amdocs - Champaign, IL   /usr/bin/perl -e '? print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");6 '1  B   ------------------------------  # Date: Wed, 21 Mar 2001 15:59:48 GMTc5 From: danco@cx48228-c.escnd1.sdca.home.com (Dan Cook)l: Subject: Re: DECnet a proprietary protocol ?  Not at all !- Message-ID: <slrn9bhkaa.rbs.danco@pebble.org>n  . On 21 Mar 2001 16:08:14 +0100, Jan Vorbrueggen8 <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> wrote:  F >All of DECnet-IV's specifications were easily available from DEC at aH >very reasonable price - so reasonable that the place I used to work forN >(a university institute running a 780 to service ~100 users intially) bought % >them without a really pressing need.   M Back in the early '80s I had a copy of the DECnet specs along with DAP specs,-K etc.  If I remember correctly, I picked them up at a DEC store while taking N an RSX class in San Jose.  They must have been inexpensive because the companyK I work for would have never paid for them.  I bought them with my own cash.oI Unfortunately, I threw them out at least ten years ago while cleaning outC accululated junk in my cubicle.y   - Dan.   ------------------------------   Date: 21 MAR 2001 16:59:57 GMT+ From: Dave Greenwood <greenwoodde@ornl.gov>:: Subject: Re: DECnet a proprietary protocol ?  Not at all !2 Message-ID: <21MAR01.16595763@feda01.fed.ornl.gov>  2 bill@triangle.cs.uofs.edu (Bill Gunshannon) wrote:  ? > I'll admit I could have been wrong.  I just remembered seeing C > somewhere that the reason it took so long to get even rudimentary D > DECNET into Linux was becuase DECNET was un-published, proprietaryE > DEC property and had to be "clean room" reverse engineered.  I know E > that I have never seen any DECNET code or specs published anywhere.a   According to the VMS faq:   B   DOC9.   Where is documentation on the DECnet Phase IV protocols?>   ------------------------------------------------------------  5   Specifications for DECnet Phase IV can be found at:t  A       http://gatekeeper.dec.com/pub/DEC/DECnet/PhaseIV/index.htmlc  1 Of course, gatekeeper.dec.com doesn't respond :-(f   Dave --------------9 Dave Greenwood                Email: Greenwoodde@ORNL.GOVkH Oak Ridge National Lab        %STD-W-DISCLAIMER, I only speak for myself   ------------------------------  % Date: Wed, 21 Mar 2001 12:34:09 -0500l- From: "Peter Weaver" <peter.weaver@stelco.ca>o: Subject: Re: DECnet a proprietary protocol ?  Not at all !4 Message-ID: <KC5u6.163323$Z2.2016886@nnrp1.uunet.ca>  8 "Dave Greenwood" <greenwoodde@ornl.gov> wrote in message, news:21MAR01.16595763@feda01.fed.ornl.gov... >... > According to the VMS faq:2 >aD >   DOC9.   Where is documentation on the DECnet Phase IV protocols?@ >   ------------------------------------------------------------ >$7 >   Specifications for DECnet Phase IV can be found at:. > C >       http://gatekeeper.dec.com/pub/DEC/DECnet/PhaseIV/index.htmlp >d3 > Of course, gatekeeper.dec.com doesn't respond :-(  >a  F This page is cached at google. Too bad you can't tell google to changeC all of the links in a document to point to the cached pages :(. TheB$ last section of the document states;  , --------------------------------------------
 The Last Wordh  ? All the protocols mentioned above (except for LAT), even though @ developed by Digital, are not proprietary to Digital. Anyone canE implement them without seeking permission from or paying royalties toe Digital.  A Eventually a number of enhancements were made to DECnet Phase IV. F These enchancements become known as DECnet Phase IV+. DECnet Phase IV+C systems were completely interoperable with DECnet Phase IV. In fact C most DECnet Phase IV systems are actually DECnet Phase IV+ systems.dB Most of the enhancements were performance improvements such as theF addition of path splitting in Routing and the addition of delayed ACKsD and a out of order cache in NSP. There was, however, one significant> new functional enhancement: the addition of proxies in SessionF Control. Note that none of the DECnet Phase IV+ changes are covered in  any of the above specifications.  D Comments or suggestions for this page are welcome and may be sent to   dnaspecs@lkg.dec.com    F ----------------------------------------------------------------------
 ----------" Last Updated on November 1st, 1994   Matt Thomas, thomas@lkg.dec.com / Network Operating Systems / Network Engineering    ------------------------------    Date: 21 Mar 2001 23:58:26 +0800, From: Paul Repacholi <prep@prep.synonet.com>: Subject: Re: DECnet a proprietary protocol ?  Not at all !- Message-ID: <87zoef9krh.fsf@prep.synonet.com>t  3 bill@triangle.cs.uofs.edu (Bill Gunshannon) writes:t  / > In article <87pufbb9jx.fsf@prep.synonet.com>, 1 >  Paul Repacholi <prep@prep.synonet.com> writes:s  F > |> Ki Research (sp?) did DECnet IV for nearly everything that opened > and shut.-  F > A quick search of the web showed numerous comments that the Spec forE > DECNET Phase IV is publicly available.  I see no mention in any doc D > found about previous versions.  Is it possible that DEC was not asF > generous with previous versions??  There is no doubt that Unix boxesC > can talk DECNET, it has been a part of Ultrix since the Ultrix-11uF > days.  Given this, if it were in fact freely avaialble, why would itC > have taken more than a decade for the first non-commercial DECNETt@ > for unix to show up and that still in a very incomplete form??  C I had (have?) a copy of all or near all of the PhIII specs. No ideam where they are at the moment...s  @ BTW, the only 'cost' to doing you own DECnet was that you had to> acknoledge that it was the DECnet orotocols, deleloped by DEC.   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  # Date: Wed, 21 Mar 2001 18:37:39 GMTh2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman): Subject: Re: DECnet a proprietary protocol ?  Not at all !/ Message-ID: <Tx6u6.11$fB6.354@news.cpqcorp.net>"  ` In article <99acn6$2oh6$1@info.cs.uofs.edu>, bill@triangle.cs.uofs.edu (Bill Gunshannon) writes:   :I just remembered seeingiB :somewhere that the reason it took so long to get even rudimentaryC :DECNET into Linux was becuase DECNET was un-published, proprietaryt? :DEC property and had to be "clean room" reverse engineered.   r  H   The "clean-room" would either involve working with engineers that had I   detailed familiarity with another vendors codebase, or it was a canard.L  K :I know that I have never seen any DECNET code or specs published anywhere.m  H   The OpenVMS FAQ has pointers to the DECnet architecture documentation,E   and has had the pointers for years.  The part numbers of the DECnetnD   architecture manuals were explicitly referenced in the old DECnet    documentation, as well.m  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Wed, 21 Mar 2001 13:53:48 +0000h  From: steven.reece@quintiles.com9 Subject: Re: DECnet a proprietary protocol ? Not at all !pH Message-ID: <OF164B0C56.EDBAB945-ON80256A16.004C29FC@qedi.quintiles.com>  G I've also heard that one of the ideas for the next generation of TCP/IPtK protocols after IPv6 was DECnet-Plus.  Exactly how I'm not sure, although I.J do know of at least one person lurking in the newsgroup that does know....   Roy Omond wrote:A >>>You hit one of my pet peeves, and my peeves didn't like it :-)b  > DECnet is not and never has been a "proprietary DEC protocol".? It is the very definition of the complete opposite.  The DECnete? protocol definitions have *always* been free for anyone to make B whatever use of them they desired.  There were several third-partyB implementations of DECnet on other platforms (I recall at least on Sun and Macintosh).<<<   ------------------------------  % Date: Wed, 21 Mar 2001 14:32:58 +00001  From: steven.reece@quintiles.com9 Subject: Re: DECnet a proprietary protocol ? Not at all ! H Message-ID: <OF557991DC.5F6460DB-ON80256A16.004FC0C0@qedi.quintiles.com>  @ It's probably more the case that nobody publishes it because theE documentation and standards are long and go down to great detail.  IPsE documentation on the other hand is perhaps not quite so detailed.....e   Bill Gunshannon commented:@ >>>I'll admit I could have been wrong.  I just remembered seeingA somewhere that the reason it took so long to get even rudimentarytB DECNET into Linux was becuase DECNET was un-published, proprietaryC DEC property and had to be "clean room" reverse engineered.  I knownF that I have never seen any DECNET code or specs published anywhere.<<<   ------------------------------  % Date: Wed, 21 Mar 2001 12:14:20 -0300C) From: fabio_compaq@ep-bc.petrobras.com.br-9 Subject: Re: DECnet a proprietary protocol ? Not at all ! L Message-ID: <OFFB72247F.A712FDCB-ON03256A16.00539E53@ep-bc.petrobras.com.br>  H I installled SunDNI a few years ago, and this Decnet for Solaris   was = so limited in functionality.    Sdsp   F=E1bio Cardoso-        B bill@triangle.cs.uofs.edu (Bill Gunshannon) em 21/03/2001 11:52:36  & Favor responder a bill@cs.scranton.edu             Info-VAX@Mvb.Saic.Com       9 Assunto: Re: DECnet a proprietary protocol ? Not at all !a    - In article <87pufbb9jx.fsf@prep.synonet.com>,</  Paul Repacholi <prep@prep.synonet.com> writes:l |>H |> Ki Research (sp?) did DECnet IV for nearly everything that opened an= dh shut.x  D A quick search of the web showed numerous comments that the Spec forC DECNET Phase IV is publicly available.  I see no mention in any doctB found about previous versions.  Is it possible that DEC was not asD generous with previous versions??  There is no doubt that Unix boxesH can talk DECNET, it has been a part of Ultrix since the Ultrix-11 days.=  H Given this, if it were in fact freely avaialble, why would it have take= nbH more than a decade for the first non-commercial DECNET for unix to show=  - up and that still in a very incomplete form??a   bill   --H Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wol= veslD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |> Scranton, Pennsylvania   |         #include <std.disclaimer.h>         =e   ------------------------------   Date: 21 Mar 2001 18:42:07 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)a9 Subject: Re: DECnet a proprietary protocol ? Not at all !a, Message-ID: <99aslv$301p$1@info.cs.uofs.edu>  H In article <OF557991DC.5F6460DB-ON80256A16.004FC0C0@qedi.quintiles.com>,#  steven.reece@quintiles.com writes:  |>  C |> It's probably more the case that nobody publishes it because theoH |> documentation and standards are long and go down to great detail.  IPH |> documentation on the other hand is perhaps not quite so detailed.....  E Yeah, it's just detailed enough to implement it in everything down touG the soda machine at CMU.  :-) (at least I think there was one at CMU!!)    bill   -- iJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   i   ------------------------------  # Date: Wed, 21 Mar 2001 16:23:59 GMT6= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)cM Subject: Dumbing down comp.os.vms with "Dumbing Down VMS with UNIX Elements?"10 Message-ID: <009F9581.E33E911F@SendSpamHere.ORG>  # Will this stupid thread ever cease?e   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COMt            nO city, n., 1. a place where trees are cut down and streets are named after them.f   ------------------------------   Date: 21 Mar 2001 16:23:22 GMT* From: bdwheele@indiana.edu (Brian Wheeler)Y Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS       EducationalProgrn2 Message-ID: <99akhq$io$1@flotsam.uits.indiana.edu>  L I was going to put a detailed rebuttal of Mr. Brass' comments, but I'm tiredK of wasting my time on someone who has no touch with reality.  I do not careNK if it means that I've "lost the argument", but I hereby invoke Godel's Law:   E 	Mr. Brass, your behavior of VMS vs Unix is frightenly reminiscent ofeD 	the Nazis in WW2 of Germans vs. the Jews.  Their reasons for hatingF 	the Jews are as irrational as your hatred (YES, HATRED) of Unix.  YouC 	insist that any exposure to anything that has been exposed to Unix C 	will contaminate VMS, much like children are worried about getting.C 	cooties from girls.  You repeatedly ignored reality when it didn'trA 	suit your argument and constantly change your "point" when it isu 	proven wrong.  M I will continue to use VMS for what is good for, as long as it is the correctlK choice.  I will use Unix for what it is good for [here Mr. Brass, I'll savetI you the trouble:  "its all crap!  Unix sucks, its good for nothing!"].  I J will continue to support the COE initiative on VMS as it might be the lastF chance for VMS to dig itself out of the niche it is being driven into.   Briane   ------------------------------    Date: 21 Mar 2001 18:42:43 -00004 From: Doc.Cypher <Use-Author-Address-Header@[127.1]>Y Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS       EducationalProgr8+ Message-ID: <ILike.BriansFlame@nowhere.nil>h  " -----BEGIN PGP SIGNED MESSAGE-----  ; On 21 Mar 2001, bdwheele@indiana.edu (Brian Wheeler) wrote:rM >I was going to put a detailed rebuttal of Mr. Brass' comments, but I'm tiredoL >of wasting my time on someone who has no touch with reality.  I do not careL >if it means that I've "lost the argument", but I hereby invoke Godel's Law: >oF >	Mr. Brass, your behavior of VMS vs Unix is frightenly reminiscent ofE >	the Nazis in WW2 of Germans vs. the Jews.  Their reasons for hatingtG >	the Jews are as irrational as your hatred (YES, HATRED) of Unix.  YoupD >	insist that any exposure to anything that has been exposed to UnixD >	will contaminate VMS, much like children are worried about gettingD >	cooties from girls.  You repeatedly ignored reality when it didn'tB >	suit your argument and constantly change your "point" when it is >	proven wrong.1 >:N >I will continue to use VMS for what is good for, as long as it is the correctL >choice.  I will use Unix for what it is good for [here Mr. Brass, I'll saveJ >you the trouble:  "its all crap!  Unix sucks, its good for nothing!"].  IK >will continue to support the COE initiative on VMS as it might be the lastiG >chance for VMS to dig itself out of the niche it is being driven into.  >t >Brian  % 'bin reading up on flaming Brian? <g>u  # You forgot the "Plonk!" on the end.      Doc.  C P.S. Read the references from my X-headers to find out how to do itr      anonymously.    -----BEGIN PGP SIGNATURE-----s Version: 2.6.2  @ iQEVAwUBOrfg8sriC3SGiziTAQFhWgf/SJuP263/kKrMGcVq7MOfxnrMI9Dd1sgc@ uyTvwClMC5r+8/fjp5rzvRwG9PHQzYl3nFrcauN3k3e3wximyjGMWo3rGaWxUDFo@ iazGkLDQRswX1kXfaoLpfa+67phFqezgHYHyVsH3WfPJsxQhvdYnjoTiAg30VTZX@ A35zxAOR28SL8lCd6r0yWqcX5DYtkoCglscmiZtdc9nIbvqBmUUTbYZbQ74+49jc@ Xogrk924fZKohHa+Jx9c9+59yLMAlI/ChHHufC2HwZAXbA9i+13UhbaEv5fP2DTf8 d+6Nuko6cYNqfU6iz+sel2brsAhvzsVuJtChw3VO05tvSM73l925uw== =rG+hg -----END PGP SIGNATURE-----m   ------------------------------   Date: 21 Mar 2001 18:49:30 GMT* From: bdwheele@indiana.edu (Brian Wheeler)Y Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS       EducationalProgrd3 Message-ID: <99at3q$1i4$1@flotsam.uits.indiana.edu>g  + In article <ILike.BriansFlame@nowhere.nil>, 7 	Doc.Cypher <Use-Author-Address-Header@[127.1]> writes:r$ > -----BEGIN PGP SIGNED MESSAGE----- > = > On 21 Mar 2001, bdwheele@indiana.edu (Brian Wheeler) wrote:fN >>I was going to put a detailed rebuttal of Mr. Brass' comments, but I'm tiredM >>of wasting my time on someone who has no touch with reality.  I do not care-M >>if it means that I've "lost the argument", but I hereby invoke Godel's Law:  >>G >>	Mr. Brass, your behavior of VMS vs Unix is frightenly reminiscent of,F >>	the Nazis in WW2 of Germans vs. the Jews.  Their reasons for hatingH >>	the Jews are as irrational as your hatred (YES, HATRED) of Unix.  YouE >>	insist that any exposure to anything that has been exposed to Unix E >>	will contaminate VMS, much like children are worried about gettingeE >>	cooties from girls.  You repeatedly ignored reality when it didn'toC >>	suit your argument and constantly change your "point" when it iss >>	proven wrong. >>O >>I will continue to use VMS for what is good for, as long as it is the correctiM >>choice.  I will use Unix for what it is good for [here Mr. Brass, I'll saveCK >>you the trouble:  "its all crap!  Unix sucks, its good for nothing!"].  ImL >>will continue to support the COE initiative on VMS as it might be the lastH >>chance for VMS to dig itself out of the niche it is being driven into. >> >>Briant > ' > 'bin reading up on flaming Brian? <g>r > % > You forgot the "Plonk!" on the end.e  * Damn, I knew there was something I forgot!   Brian    ------------------------------   Date: 21 Mar 2001 07:10:43 GMT- From: djweath@attglobal.net (Dave Weatherall)uY Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS      EducationalPrograa5 Message-ID: <DTiotGxQ0bj6-pn2-cDEtXgbGtuHz@localhost>s  ? On Tue, 20 Mar 2001 21:33:20, Ben Sego <bsego@clark.net> wrote:e   ..  ? >     If the commonly reported story is true, IBM wanted to useuB > > CP/M-86 for the PC, but DR wouldn't talk to them. So they wentC > > to Gates, who didn't even own the rights to DOS at the time. HeeB > > reportedly put them off for a day or two, made a quick deal to: > > bag the rights to DOS and then made the deal with IBM. > >r@ > >   (all of the above from a PBS series from a few years ago). > M > You forgot the part where it all happened because his mother served on somehE > charity board with an IBM exec; she got the initial meeting set up.n >   F I didn'tr see that bit the film 'Pirates of Silicon Valley'. I did see; what I thought was a pdp-8 being used to write their Basic GC complier/interpreter. Neither Jobs nor Gates comes out of the film , looking too good.y   --   Cheers - Dave.   ------------------------------    Date: 21 Mar 2001 13:10:06 -00004 From: Doc.Cypher <Use-Author-Address-Header@[127.1]>Y Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS      EducationalPrograc6 Message-ID: <20010321131006.16894.qmail@nym.alias.net>  " -----BEGIN PGP SIGNED MESSAGE-----  ? On Wed, 21 Mar 2001, Christof Brass <brass@infopuls.com> wrote:  >Paul Repacholi wrote:   <snip-snip>a  I >> I have said it before, and I will say it again. Unix is still possible-H >> the system for developing how to solve a problem, or finding out if a> >> usefull solution is possible or practicle. It is a very badG >> environment to develop a solid code base that can be ported to otherpF >> platforms. The strengths of structureless flexibility and lego likeD >> build-a-brick software that are so usefull in the beginning are a= >> foundation of sand when it is time to get a solid reliable- >> implementation done.W >h@ >The wellknow PERL trap or the Rapid Prototyping Trap. You go on@ >and on and end up with a mess of unmanagable source that nobody/ >understands and that isn't extensible anymore.m  J Here we're into the meat and potatoes of *real* software engineering. It'sE that magical idea that incompetent management get when presented withP Prototyping tools.  & The idea that Prototype = Application.  > I recall having a discussion on this subject with one of thoseJ aforementioned managers many years ago. He wanted me to dive into the codeH with no understanding of the client requirement, and stick in additionalE logic until it looked pretty. Problem was, the crap-app I was hackings? already needed 25%+ of the code hacked out to make the routinesiK understandable. Yes, this *was* on VMS. Yes, the aforementioned manager hadn5 got himself promoted away from coding for writing it.   F I still say you can't beat proper specifications - even if you have toH write them from a *throwaway* prototype because the client can't explain their requirements.   H Anyway, this isn't peculiar to VMS, it happens on every platform. That's: why so many management types like their coding cuisinarts.     Doc.   -----BEGIN PGP SIGNATURE-----M Version: 2.6.2  @ iQEVAwUBOrfg8sriC3SGiziTAQHr9Qf7B2MYiYr9/C0X2YkX6h9E0LPxEGD0Levq@ POimbSI49xRS0Pt7NB7u6ZPt/7/TAK0sd2CJDqYSmeW0mf4X36ko61J3HlnI4a5C@ hAFcw04dScJ1LLTvR5xyKeKuHJ3f84aMcsitnBfarkYz9Mpqx5h+z+ThpSufSV5u@ ZS3q7TpBtS5tPyT35Hna+PxEkGhDBaxo9JRcrCToQq3eozq3rQEqFBTZm2EfAec/@ zCRYcl5uDR0eGjg5IPm+9K0igaQmJ1ciEmsIC+TJfADv1ILxLlACn9sxPBbc4m6M8 5QdJz2l2e5D8wk5esGkWfctvUzssfQfO7chmaAXcf0sWvDqQff2ICw== =1wD4  -----END PGP SIGNATURE-----h   ------------------------------    Date: 21 Mar 2001 08:19:50 -0500- From: koehler@encompasserve.org (Bob Koehler) Y Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS      EducationalPrograw3 Message-ID: <0nbGN2dCuQfC@eisner.encompasserve.org>b  ` In article <998rei$2091$2@info.cs.uofs.edu>, bill@triangle.cs.uofs.edu (Bill Gunshannon) writes:5 > In article <wFwzVL1RVj1$@eisner.encompasserve.org>,t > ( > When was unix ever sold to the masses?  E 1990's open system hype seen in almost every computer publication you  could get your hands on.  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporationt= NASA GSFC Flight Software       | Federal Sector, Civil GrouplE                                 | please remove ".aspm" when replyingC   ------------------------------   Date: 21 Mar 2001 16:03:37 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)sY Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS      EducationalProgra-, Message-ID: <99ajcp$2rjg$2@info.cs.uofs.edu>  3 In article <0nbGN2dCuQfC@eisner.encompasserve.org>,u0  koehler@encompasserve.org (Bob Koehler) writes:c |> In article <998rei$2091$2@info.cs.uofs.edu>, bill@triangle.cs.uofs.edu (Bill Gunshannon) writes:u8 |> > In article <wFwzVL1RVj1$@eisner.encompasserve.org>, |> >  + |> > When was unix ever sold to the masses?n |> oH |> 1990's open system hype seen in almost every computer publication you |> could get your hands on.t  J What "computer publication" was targeted at the masses??  Even advertisingF rags like Datamation and Byte are not intended for the general public.K Oh, I guess there's Computer Shopper.  I wonder how much "open system hype"o
 they ran??  B The masses didn't learn about Unix till Linux came around and most$ of them still don't know about Unix.   bill   -- uJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   -   ------------------------------    Date: 21 Mar 2001 08:24:59 -0500- From: koehler@encompasserve.org (Bob Koehler)-Y Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS     EducationalProgram 3 Message-ID: <JHNHaSY0tsSG@eisner.encompasserve.org>-  ` In article <998kgk$1sss$2@info.cs.uofs.edu>, bill@triangle.cs.uofs.edu (Bill Gunshannon) writes:  E > Are you saying it is trivial to port from VMS to any other system??-E > Surely you jest!!  What other system in the world has LIB$FIND_FILEe > or LIB$DAY or LIB$GETSYI??  E Just because there are VMS specific APIs doesn't mean you have to uselC them.  Generally I find the closer you get to the hardware the lesseG portable code is (I've a tape drive manipulation program which consistsY: of 50 lines of code and 250 lines of #ifdef this-UNIX and > #ifdef that-UNIX).  Many applications don't have that problem.  H I've ported a tremendous amount of code between IBM mainframes, VMS, andG a variety of UNIX, and yes well implemented code which does not have toe- face the above issues can be trivial to port.d  G The biggest problem I faced porting code from VMS to UNIX was Suns lack H of support for the Motif window manager.  That's not a problem any more.  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporation9= NASA GSFC Flight Software       | Federal Sector, Civil GroupoE                                 | please remove ".aspm" when replying    ------------------------------  % Date: Wed, 21 Mar 2001 08:26:16 -0500t7 From: David Beatty <David.Beatty@sasSPAMITUPTHEZZZ.com>iY Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS     EducationalProgram 2 Message-ID: <5qq4Oo39s8ztBrGLJE=jNAHzuWQz@4ax.com>  A I never said it was easier nor simpler.  It's just not diffucult.,  ; Since it looks to be a slow day, I'll go ahead and whip onenB out.  The initial version won't work with ODS-5, that will require a little more work.<   David R. Beatty-  < On 21 Mar 2001 00:01:30 GMT, bill@triangle.cs.uofs.edu (Bill Gunshannon) wrote:  3 >In article <G7W3OjMzxo7TfaW0cHyNnUquOgut@4ax.com>,t; > David Beatty <David.Beatty@sasSPAMITUPTHEZZZ.com> writes:6E >|> On 20 Mar 2001 18:25:35 GMT, bdwheele@indiana.edu (Brian Wheeler) 
 >|> wrote: >|> N >|> >Out of curiosity, is there an easy way to delete a directory structure on< >|> >vms without having to run "DELETE *.*;*" over and over? >|> > 
 >|> >Brian >|> ( >|> You probably mean DELETE [...]*.*;*. >|> I >|> Depends on how you define easy.  It's not trivial, but can be done intH >|> a fairly small (recursive) command procedure -- less than 100 lines,F >|> probably closer to 50 lines.  The F$SEARCH lexical function is theI >|> key here.  I've sure that utility has been written by many folks overo >|> the years. >mH >And a 50-100 line "small (recursive) command procedure" is simpler than >e3 >     "find . -type f -exec rm -f {} \; ; rm -rf *"n >  >in what way?? >  >bill    ------------------------------  % Date: Wed, 21 Mar 2001 10:06:29 -0500s7 From: David Beatty <David.Beatty@sasSPAMITUPTHEZZZ.com> Y Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS     EducationalProgram=2 Message-ID: <bsK4OiTbOLwbNDFNwwgzl+z5qMCT@4ax.com>  6 First try.  I've not figured out an easy way to delete9 the top directory, but it's a start.  P1 is the directory|6 tree to delete (defaults to the current directory) and- P2 is any DELETE switches (defaults to none).   2 Minimal error checking and no protection checking.   $ !n$ $   On Control_Y Then Goto Abort_Rtn% $   On Error     Then Goto Ctrl_Y_Rtn- $-. $   DelTree     = F$Environment( "PROCEDURE" )* $   Depth       = F$Environmemt( "DEPTH" ) $   Exit_Status = 1D $ ! 7 $   If P1 .Eqs. "" Then P1 = F$Environment( "DEFAULT" )l $   Dir_Spec = P1 + "*.DIR"  $ !t $ Loop1:5 $       Next_Dir  = F$Search( "''Dir_Spec'" , Depth ),/ $       If Next_Dir .Eqs. "" Then Goto EndLoop1IE $       Next_Iter = P1 - "]" + "." + F$Parse( Next_Dir ,,, "NAME" ) -c 	+ "]"1 $       @'DelTree 'Next_Iter 'P2 "" "" "" "" "" 1i $       Goto Loop1 $ EndLoop1:t $   Delete'P2 'P1'*.*;*u $   Goto Exit_Rtn. $ !e $ Abort_Rtn: $   Exit_Status = $Statust $   Goto Exit_Rtno $ !u
 $ Ctrl_Y_Rtn:i $   Exit_Status = 2e $ !  $ Exit_Rtn:a $   Exit Exit_Status $ !    David R. Beattyo  0 On Wed, 21 Mar 2001 08:26:16 -0500, David Beatty+ <David.Beatty@sasSPAMITUPTHEZZZ.com> wrote:h   >oB >I never said it was easier nor simpler.  It's just not diffucult. >y< >Since it looks to be a slow day, I'll go ahead and whip oneC >out.  The initial version won't work with ODS-5, that will require  >a little more work. >B >David R. Beatty >V= >On 21 Mar 2001 00:01:30 GMT, bill@triangle.cs.uofs.edu (BillV >Gunshannon) wrote:? > 4 >>In article <G7W3OjMzxo7TfaW0cHyNnUquOgut@4ax.com>,< >> David Beatty <David.Beatty@sasSPAMITUPTHEZZZ.com> writes:F >>|> On 20 Mar 2001 18:25:35 GMT, bdwheele@indiana.edu (Brian Wheeler) >>|> wrote:  >>|> aO >>|> >Out of curiosity, is there an easy way to delete a directory structure on-= >>|> >vms without having to run "DELETE *.*;*" over and over?e >>|> > >>|> >Brianh >>|>  ) >>|> You probably mean DELETE [...]*.*;*.m >>|> wJ >>|> Depends on how you define easy.  It's not trivial, but can be done inI >>|> a fairly small (recursive) command procedure -- less than 100 lines,eG >>|> probably closer to 50 lines.  The F$SEARCH lexical function is theeJ >>|> key here.  I've sure that utility has been written by many folks over >>|> the years.s >>I >>And a 50-100 line "small (recursive) command procedure" is simpler thane >>4 >>     "find . -type f -exec rm -f {} \; ; rm -rf *" >> >>in what way??w >> >>bill   ------------------------------   Date: 21 Mar 2001 16:15:40 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon) Y Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS     EducationalProgram., Message-ID: <99ak3c$2s5g$1@info.cs.uofs.edu>  3 In article <JHNHaSY0tsSG@eisner.encompasserve.org>,n0  koehler@encompasserve.org (Bob Koehler) writes:c |> In article <998kgk$1sss$2@info.cs.uofs.edu>, bill@triangle.cs.uofs.edu (Bill Gunshannon) writes:i |> dH |> > Are you saying it is trivial to port from VMS to any other system??H |> > Surely you jest!!  What other system in the world has LIB$FIND_FILE |> > or LIB$DAY or LIB$GETSYI??a |> hH |> Just because there are VMS specific APIs doesn't mean you have to useF |> them.  Generally I find the closer you get to the hardware the lessJ |> portable code is (I've a tape drive manipulation program which consists= |> of 50 lines of code and 250 lines of #ifdef this-UNIX and nA |> #ifdef that-UNIX).  Many applications don't have that problem.N  H And this differs on VMS from Unix in what way??  Surely you don't expectI a program that manipulates hardware on an IBM390 to be easily portable tosI some other system either??  How portable a program is depends on the taskiJ being done, and the desire of the programmer to make the program portable.K Because most programmers only work in one environment, they tend to programeJ for that environment and not even be aware of others.  In the case of UnixG prgramming, this has and is continuing to change.  More an more code is J being written with an eye towards portability with machine and OS specificG code being isolated.  How much VMS code is being written today with thetF consideration that effort should be spent making it portable to no-VMS	 systems??y   |> uK |> I've ported a tremendous amount of code between IBM mainframes, VMS, andwJ |> a variety of UNIX, and yes well implemented code which does not have to0 |> face the above issues can be trivial to port.  G Nobody is arguing this, but as you yourself state, it is not feature ofi Unix that code is hard to port.    |> eJ |> The biggest problem I faced porting code from VMS to UNIX was Suns lackK |> of support for the Motif window manager.  That's not a problem any more.b  C Sounds like a VMS problem to me.  Why did they choose a proprietaryeD and expensive API when there were cheaper and even free alternatives
 available??  r   bill   -- sJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   n   ------------------------------   Date: 21 Mar 2001 16:20:24 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)HY Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS     EducationalProgram , Message-ID: <99akc8$2s5g$2@info.cs.uofs.edu>  2 In article <5qq4Oo39s8ztBrGLJE=jNAHzuWQz@4ax.com>,:  David Beatty <David.Beatty@sasSPAMITUPTHEZZZ.com> writes: |> wD |> I never said it was easier nor simpler.  It's just not diffucult. |> - |> |> >K |> >And a 50-100 line "small (recursive) command procedure" is simpler thani |> >6 |> >     "find . -type f -exec rm -f {} \; ; rm -rf *" |> > |> >in what way??   B Maybe you didn't, but the general gist of the thread here has beenC that everything is sipler on VMS as compared to Unix.  I can removemF the tree with two commands.  No 50-100 line shell script is necessary.   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   g   ------------------------------  % Date: Wed, 21 Mar 2001 12:04:38 -0500 7 From: David Beatty <David.Beatty@sasSPAMITUPTHEZZZ.com>oY Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS     EducationalProgramo2 Message-ID: <59y4OlS+WBGaeoSbDXdxjPe60yO3@4ax.com>  < On 21 Mar 2001 16:20:24 GMT, bill@triangle.cs.uofs.edu (Bill Gunshannon) wrote:  3 >In article <5qq4Oo39s8ztBrGLJE=jNAHzuWQz@4ax.com>,w; > David Beatty <David.Beatty@sasSPAMITUPTHEZZZ.com> writes:w >|> E >|> I never said it was easier nor simpler.  It's just not diffucult.r >|>  >|>s >|> >aL >|> >And a 50-100 line "small (recursive) command procedure" is simpler than >|> >i7 >|> >     "find . -type f -exec rm -f {} \; ; rm -rf *"  >|> >e >|> >in what way?? >iC >Maybe you didn't, but the general gist of the thread here has beenBD >that everything is sipler on VMS as compared to Unix.  I can removeG >the tree with two commands.  No 50-100 line shell script is necessary.- >  >bill1  F I'll grant you that.  It always depends on what you are doing, the O/SE you are doing it on, and the O/S features, strengths, and weaknesses.d  A There are some things that could be easier on VMS, like accessingt more system services from DCL.  C On the VMS side, I've found batch and print queue management (othereF than the f$getqui lexical) much easier in VMS than in either VM, MVS,  or UNIX.   Davidh   ------------------------------    Date: 22 Mar 2001 01:46:39 +0800, From: Paul Repacholi <prep@prep.synonet.com>Y Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS     EducationalProgramt- Message-ID: <87wv9j816o.fsf@prep.synonet.com>S  3 bill@triangle.cs.uofs.edu (Bill Gunshannon) writes:r  D > Maybe you didn't, but the general gist of the thread here has beenE > that everything is sipler on VMS as compared to Unix.  I can removen= > the tree with two commands.  No 50-100 line shell script ise > necessary.  E In ANY case? Even if it had hard linked cycles built in to it? If youeG start within one of those cycles? I think you will be in for a suprise.  Or several perhaps...-   -- -< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.:@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  % Date: Wed, 21 Mar 2001 11:28:20 +0000r/ From: Nigel Arnot <sysmgr@maxwell.ph.kcl.ac.uk> H Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS  ...)6 Message-ID: <009F9582.7EE9DFDC.3@maxwell.ph.kcl.ac.uk>  M > >>Out of curiosity, is there an easy way to delete a directory structure ono; > >>vms without having to run "DELETE *.*;*" over and over?t > >> > & > 1) set prot [top...]*.*;*/owner=rewdF > 2) delete [top...]*.*;*,;,;,;,;,;,;,;	! may need more depth on ODS-5  C Oh, bravo, that's the inelegant raised to art in it's own right! (IaC always just iterated DELETE [top...]*.*;* with uparrow return until  it said no files found.    > 3) set prot top.dira > 4) delete top.dir; > 0 > Was it really that bad?  I do it all the time.  J It's intrinsically inelegant, and you have to ignore  error messages aboutK non-empty directories. People with  high techno-aesthetic sensibilities are J likely to write a recursive DCL procedure instead, which will not generateE any error messages except those concerning true error conditions. Thea7 rest of us just grit our teeth and pay our dentists ;-)   
 		Nigel Arnotn- 		NRA@MAXWELL.PH.KCL.AC.UK                      7 		"In the beginning there was nothing, which exploded."a   ------------------------------    Date: 21 Mar 2001 08:27:47 -0500- From: koehler@encompasserve.org (Bob Koehler)fH Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS  ...)3 Message-ID: <7zZTdOBk5A8z@eisner.encompasserve.org>i  h In article <009F9582.7EE9DFDC.3@maxwell.ph.kcl.ac.uk>, Nigel Arnot <sysmgr@maxwell.ph.kcl.ac.uk> writes: > " > Oh, bravo, that's the inelegant    Define "elegant".  I dare you.  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporationp= NASA GSFC Flight Software       | Federal Sector, Civil GroupGE                                 | please remove ".aspm" when replying9   ------------------------------  % Date: Wed, 21 Mar 2001 14:42:27 +00006  From: steven.reece@quintiles.comG Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS ...)GH Message-ID: <OF746FD16E.752F3B0B-ON80256A16.005096D0@qedi.quintiles.com>  K One of the first "tasks" I was given at Leicester University was to write aQD deletion program to remove all of the files and directories within a@ specified tree.  I wrote it, it worked, but it still gave me theK heebeegeebees whenever I thought of using it.  I just use delete [...]*.*;*sF instead now.  A little safer as you at least get error messages on the' directories to show you your error.....9   Nigel Arnot wrote:G >>>It's intrinsically inelegant, and you have to ignore  error messagese aboutLK non-empty directories. People with  high techno-aesthetic sensibilities areeJ likely to write a recursive DCL procedure instead, which will not generateE any error messages except those concerning true error conditions. The-7 rest of us just grit our teeth and pay our dentists ;-)  <<<    ------------------------------  % Date: Wed, 21 Mar 2001 10:06:52 -0800e! From: Shane.F.Smith@Healthnet.com G Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS ...)-D Message-ID: <OF2E9A7891.42FC9C57-ON88256A16.0063720E@foundation.com>  C Easy. Natalie Imbruglia wearing a nice little black cocktail dress.    Shane           A koehler@encompasserve.org (Bob Koehler) on 03/21/2001 05:27:47 AMu   To:   Info-VAX@Mvb.Saic.ComC cc:e  H Subject:  Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS ...)    B In article <009F9582.7EE9DFDC.3@maxwell.ph.kcl.ac.uk>, Nigel Arnot% <sysmgr@maxwell.ph.kcl.ac.uk> writes:  >i! > Oh, bravo, that's the inelegantr   Define "elegant".  I dare you.  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporation = NASA GSFC Flight Software       | Federal Sector, Civil GroupnE                                 | please remove ".aspm" when replyingn   ------------------------------  % Date: Wed, 21 Mar 2001 06:00:17 -0500n  From: John Santos <JOHN@egh.com>Y Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS EducationalProgram) Pro6 Message-ID: <1010321054351.21646A-100000@Ives.egh.com>  " On 20 Mar 2001, Bob Koehler wrote:   > In article <rdeininger-1903011143300001@user-2ive7gk.dialup.mindspring.com>, rdeininger@mindspring.com (Robert Deininger) writes:  > > @ > > What does changing 'p' to 'l' have to do with ps or opendir? > >  >  > ps --> ls2 > 7 > ls can be written using opendir() et. al., ps cannot.u > H > ----------------------------------------------------------------------A > Bob Koehler                     | Computer Sciences Corporation ? > NASA GSFC Flight Software       | Federal Sector, Civil GroupsG >                                 | please remove ".aspm" when replyingf  E ... which nicely illustrates the Unix shell problem that Christof has 1 been ranting about for the last several weeks ;-)E  B Actually, I think this is not really a shell problem as much as itE is a shell vocabulary problem.  The shell doesn't force "show system"tC to be "ps" (process status?) or "directory" to be "ls" (list files?i! then why isn't it "lf" or "lst"?)n  F (I'm not sure, but I think that having multiword commands, like "show I system", would be difficult to implement in most Unix shells, since they eJ seem to treat the first token as either a built-in command or the name of F a program to execute.  Unless all the "show" commands were executed by a single program, etc.)   E You could, if you were twisted enough, re-write all the VMS utilitiestA and their .CLD files to use an equally arcane set of commands and-A qualifiers.  It's just that from day one, command consistency waseD a design goal of DCL, but the unix shells had no design goals.  They just grew, randomly.  C DCL is far from perfect (the quote-quote-quote-quote-quote syndrom,mB lack of sophisticated control structures, etc.) but most of us VMSB users are used to its idiosyncracies.  I'm sure "cat" and "ls" are@ second nature to anyone who has been forced to use Unix for more than a few months.  A The DCL commands are all English words, whereas the Unix commandsu@ usually look like random strings.  Clearly, the DCL commands are< easier for people who know English, but is it any benefit to> use non-random strings for non-English speakers?  I'm sure the@ consistency must help - once you know what "/confirm" means, youA also know it works anywhere confirmation makes sense and has beens implemented.  9 Rats.  I swore I was going to stop reading this thread...a   -- t John Santost Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------   Date: 21 Mar 2001 15:59:56 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)oY Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS EducationalProgram) Pre, Message-ID: <99aj5s$2rjg$1@info.cs.uofs.edu>  6 In article <1010321054351.21646A-100000@Ives.egh.com>,#  John Santos <JOHN@egh.com> writes:b |> e |>  E |> Actually, I think this is not really a shell problem as much as it,H |> is a shell vocabulary problem.  The shell doesn't force "show system"  |> to be "ps" (process status?)   - The shell doesn't force anything.  Does DCL??   F |>                              or "directory" to be "ls" (list files? |> then why isn't it "lf"   E For the same reson it isn't "dir". "ls" was thet name the originators  decided on.-  $ |>                        or "lst"?)  A Other than the reason given above, and considering the reason forZA the terse command set in the first place, the extra letter has no:
 advantage.   |> iI |> (I'm not sure, but I think that having multiword commands, like "show rA |> system", would be difficult to implement in most Unix shells, b  C Does DCL do multiword commands??  "SHOW" is the command on VMS and  B everything that follows are options.  Why would you expect Unix toA be any different??  How hard would it be to implement under DCL??   M |>                                                                since they  M |> seem to treat the first token as either a built-in command or the name of "I |> a program to execute.  Unless all the "show" commands were executed byn |> a single program, etc.)  1 You mean like they are on VMS??  [SYSEXE]SHOW.EXEr   |> eH |> You could, if you were twisted enough, re-write all the VMS utilitiesD |> and their .CLD files to use an equally arcane set of commands and |> qualifiers.    E You could, if you were twisted enough, re-write all the Unix commands F to use an equally verbose set of commands and qualifiers.  As a matterE of fact, many of the more recent incantations do use more verbose andnF human readable options.  And re-writting the commands themselves wouldG be trivial.  "mv /bin/ls /bin/directory" fixes that one and a judicious I use of links (ie. "ln directory dir", etc) gives you abbreviated commandssF as well.  Not that I would reccommend this approach, but it does show ( how easily customizable Unix actuall is.  D |>              It's just that from day one, command consistency wasG |> a design goal of DCL, but the unix shells had no design goals.  Theyi |> just grew, randomly.B  : Absurd, but then, most of the drivel against unix here is.D First, the commands (ie. ls, ps, rm, ln, mv, etc.) are not a part of? the shell int he first place any more than SHOW is part of DCL.aB Second, the design goals of Unix were clearly known (and have beenA written about in numerous papers of the era) by the implementors. E Third, they accomplished exactly what they were looking to do and didaG this with far greater success than even they had originally envisioned.n  D There is a good reason behind the terse command set and the numerousC small utilities that make up Unix.  One of them is not specifically F necessary any more, but has remained because the users are comfortableE with it and it works.  But in line with the concept behind the second F part the system is totally customizable to meet the needs of the user.E The machine should bend to the users needs, not the other way around.{   |> rF |> DCL is far from perfect (the quote-quote-quote-quote-quote syndrom,E |> lack of sophisticated control structures, etc.) but most of us VMS ) |> users are used to its idiosyncracies. w   As is equally true for Unix.  OD |>                                       I'm sure "cat" and "ls" areC |> second nature to anyone who has been forced to use Unix for morel |> than a few months.o  > I have never seen anyone hold a gun to someone's head and make> them use Unix.  Most of the people I have known become totally@ comfortable with Unix in a very short time.  let's look at where> I work now.  When i arrived, there was no Unix.  0, Zip. Nada.@ The first Unix box on campus was a DECStation on my desk, chosenA because i was brought in here to network the campus and there aret< not a lot of networking tools for IBM 3270's or DEC VT100's.@ (A 4331 and a VAX were the Administrative and Academic computers at that time.)  @ The IBM was the first to go and Administrative work was moved toC a second VAX.  So, VMS was firmly entrenched and Unix was virtually @ non-existant.  The first Unix boxes in general use were 5 SUN3'sC in the CS Department.  Shortly thereafter, the CS Department boughtt? ~40 Sparcs.  Hmmm.  Why would people stop using VMS in favor ofeA something as hard to use as Unix??  2 years later I joined the CS @ Department.  Within a year, the CS Department was independant in> it's use of the network. we had our own Email server, our own > WebServer, our own dial-up access for professors at home, etc.? The rest of th campus still relied on a purely text based emailnA system on the VAX (actually, it may have become an Alpha by then,f? but it was VMS just the same.)  They had only text based dialupy? access where we had support for subnets in the professors homeso3 (this before the idea of an ISP was even known!!)  a And so on, and so on.o< Nobody had any problem adapting.  Everyone changed willingly< becuase they had the option not to as the VMS machine stayed< there. (Actually, everyone didn't.  Two Professors still use the VMS box for their mail.)  = And the rest of the campus??  They too have all but abandoned > VMS.  The box is there, but use has dropped tot he point where< I was told I will need to take it over shortly or it will be? shut down completely. (We do still need it for COBOL and ORACLEh> for classes, however, the Prof who teaches databses is already? talking about dropping Oracle and using Postgres, which we have > running on a Unix box within the department.  That leaves only@ COBOL.  hardly justification for continuing to run the machine.)     |> f+ |> The DCL commands are all English words, t  C We've been through this already.  While english words are possible,-A few experienced VMS people actually type out the whole word, thuslC choosing to use terms nearly as terse and obscure as those on Unix.M  E |>                                          whereas the Unix commandsD& |> usually look like random strings.    @ Only to those who have not taken the time to learn the jargon of@ their business.  How many people can understand what two doctorsD talking shop to each other are saying??  Or two electrical engineers discussing a circuit??  C |>                                    Clearly, the DCL commands ared? |> easier for people who know English, but is it any benefit to 4 |> use non-random strings for non-English speakers?   ? As I have pointed out, it is trivial to make the Unix commands n? English-like, or for that matter, German-like, or Spanish-like,h& (within the limits of character sets).   bill     --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   -   ------------------------------    Date: 21 Mar 2001 18:48:59 +0100G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> Y Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS EducationalProgram) PreH Message-ID: <y4snk7rp10.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  3 bill@triangle.cs.uofs.edu (Bill Gunshannon) writes:O  E > Does DCL do multiword commands??  "SHOW" is the command on VMS and T( > everything that follows are options.    F Nope. Just shows how much you know about the workings of DCL's command! parsing. Do a VERB SHOW sometime.    	Jan   ------------------------------  % Date: Wed, 21 Mar 2001 09:14:57 -0600e+ From: Christopher Smith <csmith@amdocs.com> Y Subject: RE: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS EducationalProgram) Pr L Message-ID: <3B55D7F383B0D31197D9009027541CBF0BDD54B5@cmiexch1.cmi.itds.com>   > -----Original Message-----2 > From: Christof Brass [mailto:brass@infopuls.com]  < > But why the hell isn't it possible to simply pipe the find > result into rm?a  H It is possible to do, depending on your shell, of course, something like this:d  ! rm -f `find . -name foo\* -print`o  H This will have the desired affect, I believe.  The simple answer for theE reason that you can't pipe things into rm is that rm doesn't read its K standard input looking for that kind of thing. :)  Why doesn't it?  I don't-L know.  Could be because they waited until they were about to fill a disk up,$ and threw it together in a hurry. :)   Regards,   Chriss  ! Christopher Smith, Perl Developer  Amdocs - Champaign, IL   /usr/bin/perl -e '? print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");b 'o   ------------------------------  % Date: Wed, 21 Mar 2001 09:33:00 -0600n+ From: Christopher Smith <csmith@amdocs.com>nY Subject: RE: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS EducationalProgram) PrnL Message-ID: <3B55D7F383B0D31197D9009027541CBF0BDD54B6@cmiexch1.cmi.itds.com>   > -----Original Message-----2 > From: Christof Brass [mailto:brass@infopuls.com]  A > Point 3 is completely wrong. BTW have you ever tried to compile A > Maxwell, the text processor? Do that and come back and tell us! > > And there is one additional thing wrong with point 3: beeing> > compatible with UNIX isn't a point in favour of Linux at all< > because the Linux developers and users give a shit on thatB > because they never ever think a second about UNIX which in their< > opions is a crap, old, shitty and very expensive OS, which< > exists in many incompatible flavours. Why would they care?  L That's not the impression I get from most linux users -- which ones have youH been talking to? :)  I've even managed to get several very adamant linux9 fans interested in -- or at least appreciative of -- VMS.,   Regards,   Chrise  ! Christopher Smith, Perl Developer  Amdocs - Champaign, IL   /usr/bin/perl -e '? print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");n '    ------------------------------   Date: 21 Mar 2001 17:14:33 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)tY Subject: RE: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS EducationalProgram) Pr-, Message-ID: <99anhp$2svg$2@info.cs.uofs.edu>  L In article <3B55D7F383B0D31197D9009027541CBF0BDD54B5@cmiexch1.cmi.itds.com>,.  Christopher Smith <csmith@amdocs.com> writes: |> > -----Original Message-----d5 |> > From: Christof Brass [mailto:brass@infopuls.com]r |> i? |> > But why the hell isn't it possible to simply pipe the find  |> > result into rm? |> nK |> It is possible to do, depending on your shell, of course, something like  |> this: |> 0$ |> rm -f `find . -name foo\* -print` |> v2 |> This will have the desired affect, I believe.    G A more important question is why would you want to make "find" generaterG output to pipe into "rm" when "find" is already capable of calling "rm".@ directly with the filenames as a parameter in the proper place??  K |>                                                The simple answer for the-H |> reason that you can't pipe things into rm is that rm doesn't read itsN |> standard input looking for that kind of thing. :)  Why doesn't it?  I don'tO |> know.  Could be because they waited until they were about to fill a disk up, ' |> and threw it together in a hurry. :)-  @ Right, and my VAXStation has no sound support because all of the DEC engineers were deaf. :-)  F Lots of Unix commands don't take parameter information from the stdin.5 Usually because it makes no sense to do it that way. n  ? Question: Can you pipe a list of filenames into DELETE on VMS??v   bill   -- rJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   t   ------------------------------  % Date: Wed, 21 Mar 2001 11:45:25 -0600 + From: Christopher Smith <csmith@amdocs.com>'Y Subject: RE: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS EducationalProgram) PrtL Message-ID: <3B55D7F383B0D31197D9009027541CBF0BDD54BC@cmiexch1.cmi.itds.com>   > -----Original Message-----D > From: bill@triangle.cs.uofs.edu [mailto:bill@triangle.cs.uofs.edu]   > doesn't it?  I don't? > |> know.  Could be because they waited until they were about   > to fill a disk up,) > |> and threw it together in a hurry. :)6  B > Right, and my VAXStation has no sound support because all of the > DEC engineers were deaf. :-)  L Well, depending on the VAXStation, possibly cause they couldn't hear it over the fans anyway. :)a  H > Lots of Unix commands don't take parameter information from the stdin.7 > Usually because it makes no sense to do it that way.    J You know, I was thinking the same thing, but I stopped after I came to theJ conclusion that it does make sense.  Filenames are really the only type ofH information rm deals with, so there's not a chance for confusion.  I canE remember a few times where I've wanted to take the contents of a file K (install logs, sometimes, slightly edited), and remove them all.  Of course-L at that point you end up doing the aforementioned rm `cat this-file`, ratherJ than cat this-file|rm, or even rm <<this-file.  And why couldn't you do it the other way?  I If it were a command like "ln," on the other hand, you certainly wouldn't G want to do the same with that. :)  well... maybe if you had _two_ stdin H streams, which I suppose would be possible given some radical re-design.   Regards,   Chriss  ! Christopher Smith, Perl DeveloperI Amdocs - Champaign, IL   /usr/bin/perl -e '? print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");a '.   ------------------------------  % Date: Wed, 21 Mar 2001 10:09:05 +0000 ) From: Christof Brass <brass@infopuls.com> Y Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMSEducationalProgram) Pro , Message-ID: <3AB87DC1.9A6D1749@infopuls.com>   Paul Repacholi wrote:- > . > bdwheele@indiana.edu (Brian Wheeler) writes: > G > > > Okay. But what is the best configuration to avoid artificial linee: > > > breaks leaving sometimes only a few words on a line? >  > > Beats me, I use emacs. > 8 > :) If you want to avoid the few words on a line, use aD > type-setter. Any other method won't work, and throwing the problemF > over the wall as you are doing with unbroken lines is worse. ALL theH > line breaks are artificial for most readers, and for many, they are in" > the middle of the words as well.  < I hoped for a solution where like in Navigator the lines are@ broken according to the window width and of course not within an  word instead on word bounderies.  F > > > This is true but there are many quirks necessary to have a largeD > > > app compile on different UNIX flavours. Even well designed andA > > > well implemented C apps (oxymoron, I know) need substantialeH > > > maintenance to be portable in that sense. Have you ever read theseH > > > C header files with these endless #ifdefs and #defines coping withB > > > the small but tedious (and from my point of view unecessary)( > > > differences? I can only say: ugly! > ; > > Ugly, perhaps.  But C isn't the only language for apps.C > H > But the lack of a common call standard on any of the unixes means eachC > lang is an isolated island. Plus, if you go the route of multiples7 > languages, you end up with the N^2 interface problem.e  $ Exactly! BTW the CORBA problem also.  C > There are some well engineered apps on unix. What I find massivlyEG > depressing is the order of effort to have them portable at all acrosssH > unix variants, and the mind-numbing errors people who have done it hadF > to overcome to get resonable results. See David Tilbrooks papers for
 > an example.o > G > > >> Don't get me wrong, there are some valid criticisms, but many ofwG > > >> them are toolset issues.  Want a standard user experience acrosssH > > >> all unixes?  Install the GNU tools everywhere...magically most of+ > > >> the complaints addressed are solved.0 > D > Funny, thats what I do with unix boxes I get. Even HPUX is livable > (almost) after that is done. > E > > The kernel structure of Unix is easier to understand, at least onw > > the conceptual level.r > H > Do you think so? This I would like to see if you mean the internals of> > unix. The external face is easy, there is very little of it.  	 My point.>  H > I have said it before, and I will say it again. Unix is still possibleG > the system for developing how to solve a problem, or finding out if as= > usefull solution is possible or practicle. It is a very badcF > environment to develop a solid code base that can be ported to otherE > platforms. The strengths of structureless flexibility and lego likerC > build-a-brick software that are so usefull in the beginning are a-< > foundation of sand when it is time to get a solid reliable > implementation done.  ? The wellknow PERL trap or the Rapid Prototyping Trap. You go on ? and on and end up with a mess of unmanagable source that nobodyu. understands and that isn't extensible anymore.   > --> > Paul Repacholi                               1 Crescent Rd.,9 > +61 (08) 9257-1001                           Kalamunda.nB >                                              West Australia 60760 > Raw, Cooked or Well-done, it's all half baked.   ------------------------------  % Date: Wed, 21 Mar 2001 09:57:33 +0000 ) From: Christof Brass <brass@infopuls.com>sY Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMSEducationalProgram) Pro , Message-ID: <3AB87B0D.53DD5336@infopuls.com>   Brian Wheeler wrote: > . > In article <3AB6C610.E03722F7@infopuls.com>,5 >         Christof Brass <brass@infopuls.com> writes:w > > Brian Wheeler wrote:
 > > [SNIP]C > >> >> So where is VMS the right tool and unix not the right tool?n > >> >X > >> > Where you want to a stable and easy to administer environment with good VMS apps. > >>R > >> Unix is as easy to administer.  You have not shown any cases where it is not. > >m? > > Statistics, independent analysis and experience showed thato@ > > equally good people are about twice as efficient with VMS as > > with UNIX. > N > care to point to these analysis?  I submit to you that 80% of all statistics > on usenet are made up.  6 This is one of the 20%. I'll post them if I meet them.   > >> What VMS apps?o > >oA > > Most prominent examples are Database centered apps, real-time D > > production and management apps (BWM, Bayer, BASF, power stationsB > > all around the world) and financial transactional systems likeC > > SWX Swiss Exchange, DBAG Frankfurt, Eurex, CBOT. Any questions?i > N > Yeah, where are the apps for 99% of the potiential VMS customers?  These areO > fine examples of vertical markets, but they are worthless from the standpointx > of general acceptance.  7 No OS is generaly accepted. What does this mean anyway?h  P > >> >> So then, unix isn't crap if it does the job.  That's what its all about. > >> >D > >> > I don't think so. I'm also interested what is under the hood. > >> > >> Fair enough.n > >nB > > BTW this open the hood attitude is a recent outcome of the badD > > experiences companies made with poorly crafted but running apps.D > > If your programs don't have structure and don't obey good design> > > rules the maintenance and the future development will be a > > nightmare. >  > <sarcasm>hI > Oh, and nobody has ever written a bad app for VMS, they always stick tot > unix for that! > </sarcasm> > I > Bad apps (as well as good ones) can appear on any platform at any time.p  ; Compare M$DOS, Windoze Con (consumer versions), Windoze Buse: (business versions) and MacOS. We're not talking about the exceptions from the rule!   K > I submit to you that while there are some crappy apps for unix, there areoI > more good ones for unix than there are for VMS, solely because there isa# > more software available for unix.m  > I don't buy that shit. Where are the good ones for UNIX? Don't mention Navigator!  
 > > [SNIP] > >i > >> > > >> > Let's see when and if it's available to the SW vendors. I'm not sure if I understood Fred Kleinsorge correctly. It seems that for some time the UNIX API won't in the core VMS. > >>A > >> Agreed.  Then we can argue about if its a good or bad thing.n > >lA > > Unfortunately if it turns out to be a major step in the wrongsB > > direction the money has already been spent. That's why I write > > against it now.t > L > Well, if its already spent, then your bitching isn't going to do anything.  < Then it will have been spent - after delivery. Do you really read what is written/?  
 > > [SNIP] > >tL > >> >> Yeah, and apparently unlike you, I'm not ignorant on how it became aS > >> >> standard.  It was forcefully bundled with mass market OEM machines.  MS wasnU > >> >> weilding monopoly power in the DOS market to force windows onto users whetheroT > >> >> they wanted it or not.  When enough installs were out there, *ONLY THEN* wasU > >> >> it a successful platform to write to.  VMS is very different in this respect:eL > >> >> realistically one vendor for hardware, very small market share, etc. > >> > > >> > While this paragraph is also what I think is the truth it sadly enough isn't related to my argument and doesn't help you in defending your position. Remember - we were talking about the word "proprietary". > >>P > >> Yes.  You were arguing that proprietary doesn't matter if you've got market+ > >> share.  VMS doesn't have market share.r > >d= > > You are missing the point. No OS had the monopol from the D > > beginning besides M$DOS. There isn't any law that getting market > > share is forbidden.s > H > MSDOS did _NOT_ have a monopoly at the beginning either.  CPM-86 was aL > viable alternative, and was readily available (though more expensive).  ItP > only wasn't as popular because of bundling agreements, which is shady at best. > P > YOU are missing the point:  VMS cannot continue to do its own thing and expectN > to magically come out of the dark and be everyone's favorite OS.  A catalystJ > is required.  This catalyst may very well be the COE initiative.  It may2 > not.  But not trying it isn't going to help any.  : You obviously didn't read the sound analysis that has been= performed in this NG. Otherwise you would have refrained fromn: stating that this is the one and only cure for VMS to gain< market share. In fact it is the wrong way as we now can very safely say.i  
 > > [SNIP] > >h > >> Beats me, I use emacs.m > >lB > > Haven't tried that on email although I know you can all within? > > emacs and I have even emacs on VMS. You can even browse Web ? > > pages within emacs. Is this true? And if, how does it work?  > H > yes, emacs uses a w3 module which lets you view web pages.  I just useD > emacs for editting from knews (my newsreader) and elm (my mailer).  ? I'll give it a try. To be better than Navigator isn't that higha a goal.i  
 > > [SNIP] > >sP > >> >> We agree, but only to an extent.  VMS is proprietary in that there is noQ > >> >> independent (i.e. not controlled by compaq) which says "this is VMS".  ItkO > >> >> is available from only one vendor, and is not compatible with any othertD > >> >> system at the source level (except for very trivial things). > >> >> >> > Agreed. But I stay to what I wrote: this isn't a technical argument. And it's per se not even an economical argument as we can see with Micro$oft and to some extend with Apple which uses its beeing different (read proprietary) to a certain degree as marketing.  > >>O > >> MS forced marketshare to create a de-facto standard.  VMS doesn't have thelP > >> marketshare to do this.  Apple had obvious benefits over DOS, which negatedR > >> its proprietary disadvantage.  VMS may have advantages over Unix, but they're@ > >> not "obvious", so its proprietary nature is a disadvantage. > >n > > Too simple, I don't buy it.J > B > What about Occam's razor, you were so keen on promoting earlier?  > Make it as simple as possible *but not simpler*. Ever heard of@ marketing to explain to people the avantages of one system? Your= simplistic view of the economical world isn't a great help in5 doing some sound analysis.    > > This is rather a question ofC > > perception, marketing and management culture (which isn't therei8 > > anymore unfortunately). To me and every educated andA > > knowledgable engineer who studied both systems the advantagese< > > are more obvious than the advantages of MacOS < 10 above > > Windoze. >  > <dripping with sarcasm!>H > Yeah, cooperative multitasking is *FAR* superior to preemptive.  Its aI > wonder that VMS doesn't use it!  Heck, fixed run-time memory limits areWE > a godsend compared to dynamically being able to allocate as much asy > is required at run time. > </dripping with sarcasm> >  > You are kidding, right?i  : No, but I don't have time to teach you all the problems of; preemptive multitasking and the other wrong assumptions andr@ conclusions. Preemptive mt has several advantages but also sever@ disadvantages wrt resource usage and flexibility. You'll find it out.  > >> > This is true but there are many quirks necessary to have a large app compile on different UNIX flavours. Even well designed and well implemented C apps (oxymoron, I know) need substantial maintenance to be portable in that sense. Have you ever read these C header files with these endless #ifdefs and #defines coping with the small but tedious (and from my point of view unecessary) differences? I can only say: ugly! > >>< > >> Ugly, perhaps.  But C isn't the only language for apps. > >tC > > If you had read some recent posts of mine you would know that IdD > > try to convince the programmers to avoid C because it's the same< > > crap as UNIX - both are married. The UNIX and C attitudeD > > complement each other in one of the fruitfulst ways resulting in: > > very unreliable UNIX variants and apps on top of them. > >u@ > > C is used too often. UNIX is mainly implemented in C. Do you2 > > need any other reasons to stay away from UNIX? > J > That's not a reason, that's a bias.   BLISS is crap, do I need any otherQ > reason to avoid VMS?  Macro-32 promotes poor programming and should be avoided.\E > Are these reasons?  No.  No more than your C==Unix==crap arguments.-  = If BLISS were crap we need reasons why to trust VMS' quality.l@ Unfortunately the UNIX crap has proven its unreliability (a.s.o.= I don't enumerate all the nasty attributes of UNIX crap), andP> moreover the UNIX crap apps have proven their shittyness. With@ VMS the opposite is the case. So we need reasons why to use UNIX< crap and its crap apps despite all the facts of its built-in@ crap base. Take some time a read the analysis and proof of C/C++: crap within cov. There are even well written books on that@ topic. I never encountered something like that against BLISS and? VMS. But I'm open to see what you have to show. So far the caset? against UNIX and C/C++ is closen - technically they are extremei@ crap. There is no need to continue this part of the discussion -@ this has BTW nothing to do with bias or hate. What is still open0 - as I mentioned it - a sound analysis of BLISS.  L > You keep repeating that "any knowledgable engineer" should determine theseL > things...yet the bulk of your arguments are based on bias and anything but/ > facts.   How do you explain this discrepency?t  : Unfortunately this is your problem of misunderstanding and perspective. See above.    > >> >> >> > UNIX and Windoze are another form of beeing proprietary with UNIX having the disadvantage of never beeing the same if you change the vendor.S
 > >> >> >>R > >> >> >> Are they as great as when you move from VMS to anything else?  Not evenU > >> >> >> close.  The difference between unixes is trivial at best for the end user,s= > >> >> >> and only require a bit of effort for the sysadmin.n	 > >> >> >E > >> >> > Yes - the differences between the UNIXes are much smaller than between VMS and UNIX. What a surprise! VMS isn't UNIX? Why isn't UNIX implementing the Windoze API. There is the vast majority of apps! > >> >>nS > >> >> The WINE project is implementing the windows api on unix. Corel has used it-S > >> >> to port several applications to linux via a recompile.  Also, products suchm/ > >> >> as WindU has been doing this for years.n > >w< > > Corel messed several things up. Do they still exist? ;-)? > > I think Corel isn't a good example of putting things right.O > J > Corel may be a business disaster area, but they do some technical things > right.  < Yup. Especially Corel Draw and I frankly admit that business> desaster doesn't imply technical desaster (see VMS!). But WINE@ as I coincidentally just read in one of the best EDP magazins is> still in an immature state. Basically to come back on track it@ burns down to the point why WINE/WABI isn't a UNIX crap standard? (sorry of using this oxymoron by combining the words "standard" > and "UNIX") while OTOH it is expected to be applaused that VMS@ is ruined by imlementing the UNIX crap API within the kernel. To< point on some of the weaknesses of your argumentation: every? educated and knowledgable engineer should be able to understand @ the fundamental difference between the "UNIX crap on VMS COE put9 it in the kernel of VMS" shit and the layered approach oft
 WINE/WABI.  > >> > I know WINE and WABI - good approaches but there is something wrong. Both never took off. It's similar to the FX32! although the FX32! was even much better. It seems to me very interesting that the emulation approach isn't very popular at the moment. Instead the trend is towards using universal library APIs like Qt and providing apps on all major platforms (no, not including VMS).0 > >> > Have you good experiences with WINE apps? > >>R > >> Yes.  They're not perfect, but they're not awful, either.  WINE also compilesR > >> to native libraries, so when you write a win api app, you can recompile it onJ > >> any platform that WINE supports...so its not just an emulation issue. > >aA > > All projects *I* know about of using WINE didn't go well. Anyn > > counterexamples around?m > I > I gave some, with the corel applications.  Not to mention some in houses > stuff we have.  : Honestly I never heard of anyone using this Corel stuff on8 Linux. And honestly your site has basically not an, uhm, outstanding reputation.u  
 > > [SNIP] > >iJ > >> > I don't fight for these small differences. This example is for you. > >> > But I'm not sure if your statistics is correct. I use ps -aux on Linux which doesn't work on the Solaris version I'm using (2.6). I studied the man page for ps on Solaris and found that the option set is *very* different from the Linux one.r > >>J > >> Its a BSD vs SYSV issue.  Linux allows either set of options, but theS > >> 'default' solaris one allows only one set.  If you put /usr/ucb (The BSD stylecI > >> utils) in the head of your path, ps -aux works just fine on solaris.d > >dD > > Thanks, I'll try that. But honestly this isn't a point in favour > > of UNIX crap.p > 
 > Why not?  @ Because it means that even the basics are different and you have8 to try to adjust your system to let it operate in a more8 familiar way if you switch from one UNIX crap flavour to9 another. Are you sure this adjustment can be done always?f  P > >> >> Sure, they have different 'exotic' features, but lets face it, the basic" > >> >> functionality is the same. > >> >f > >> > This is true. But I think this isn't very helpful as most people need additional functionality. > >> >M > >> >> For find, there are more differences, but the basic usage is the sameh> > >> >> regardless of platform:  find <dir> -name "<filename>" > >> >'> >> > Honestly the basic functionality isn't very helpful. Especially the combination with grep is often needed. How can you case insensitively search on Solaris? Linux: find <dir> -iname '<filename>'. What about the time window -mmin <time1> -mmax <time2>? How is this accomplished on Solaris?eJ > >> > Honestly the quotation needed to make find work with grep is silly. > >> > And why do I have to use this -exec option at all. Why can't I use pipeing the list of file names found by "find" into grep???  > >>
 > >> You can:t& > >>         find / | grep -i filename > >> should work just fine.t > >rC > > I don't get it. What does "filename" denote? If you had writtent; > > <filename> this would have made sense from the point of F > > presenting the command but not from its usage because the list of B > > filenames should be provided by find. Anyway this doesn't work@ > > and you should have known this because there is a simple but/ > > stupid reason because this won't work ever. ? > > The solution to this problem is left as an exercise for theo > > reader.  > J > Bitching about typos doesn't validate your point.  What is this "stupid" > reason?  It works just fine:  , I'm not bitching about the typo - see below!  , > [root@wombat /]# find / | grep -i REBOOT.h > /usr/include/sys/reboot.hu/ > /usr/local/include/oskit/freebsd/sys/reboot.hr. > /usr/src/linux-2.2.14/include/linux/reboot.h. > /usr/src/linux-2.2.17/include/linux/reboot.h0 > /usr/i386-glibc20-linux/include/linux/reboot.h. > /usr/i386-glibc20-linux/include/sys/reboot.h > 3 > Testing piping of find into grep on other unixes:g >         HPUX:   yepu >         Solaris: yep >         Digital Unix:  yep( > I'm going to have to say you're wrong.  > You completely missed the point and that's obvious because you= thought that I were bitching about a typo while in fact I was ? complaining of a potential misunderstanding of yours which came   up to its full clarity only now.  8 Your command example works but does something completely< different than the "-exec grep" combination. If your example? shows how to search case insensitively for a file than I refusee; that because the Linux example didn't need the pipeing intor@ grep. If you carefully analyse this example you'll see that this7 is another example of the stupid UNIX philosophy of not ? implementing things according to concepts/standards/regularity.l  R > >> >> This works the same on all unixes I've dealt with.  Again, its an issue ofQ > >> >> 'exotic' features...sort of like when you use VMS 5.5 and bitch about notr > >> >> having 'PIPE'. > >> >& > >> > Sorry? Do you really mean that? > >> > >> :)  Sort of.e > >e
 > > [SNIP] > >  > >> > I don't agree besides the last statement which is awfully true and reveals at the same time the whole UNIX business pitty.t > >> > What makes you think that it's anything harder to make Windoze clone than a UNIX clone? And isn't VMS certified as "open"? What does this mean? Doesn't it mean that the interfaces are well documented, stable and published?a > >nS > >> Take a quick peek at the wine archives and see how often the documentation foraQ > >> windows is wrong, incomplete, or misleading.  Windows defines itself as OpenA6 > >> as well...these days 'open' doesn't mean a thing. > > C > > Stop that. I read that VMS has been certified by an independentk > > organisation.o > G > I've read lots of things too, but that doesn't make them true withoutc
 > references.e  < This has been contributed by another poster and there wasn't anything against that.  6 > > And you probably know that the incongruence of theC > > Windoze documentation and implementation is a quality assurance D > > problem with $$$ (should denote M$). Do you remember the russianB > > DOS clone? Better, faster, more stable. Implemented whithout aC > > bit of official documentation others than the readily available  > > one. > M > DOS isn't Windows.  DOS has its share of undocumented features, but there'smO > a pile more in windows, and its not just an "oops we forgot to document this" I > issue from MS.  Windows is a moving target, which makes it even harder.e  = Of course you have to concentrate on one version and fetchingo- some source through the Internet may help ;-)o  
 > > [SNIP] > >e> >> > Don't take it personal. You aren't UNIX. If I criticise UNIX it's because it sucks and I've lost too much time in believing its simplistic anti-philosophy. If I insult you than because you do so to me in constantly avoiding arguing to the point. I offered to agree on practical testing - no reaction. Do you know any board of engineers, independant from UNIX and VMS, to which we could presented our arguments?r > >>S > >> Ok, fair enough.  I've been abusive, and I'll stop it.  It seemed the only wayU3 > >> to reply to your opinions of anything non-VMS.h > > A > > I'm against UNIX and C/C++ because I know it too well and I'mlF > > sure the evolution is far beyound these archaic systems/languages. > H > They seem to be archaic only in your point of view.  A hammer is quiteA > archaic, yet it works very well for the job it is designed for.h  > Very good point. Unfortunately not in favour of UNIX crap. Try; to create a microchip with a hammer. I like it very much to @ agree to you that UNIX crap is the SW/OS equivalent to a hammer.< Please don't use this archaic tool for solving real problems nowadays. Thank you!  - > > I'm not against BeOS (I don't know it too C > > well and I don't like that they used C++ for implementation but % > > they offer some very good ideas),N > J > Yep, and its posix compliant which means that it has a bunch of software > from the get-go. > / > > not against MacOS from the UI point of view  > L > Yet you rag on OS X, which is still (pretty much) the same user interface. > 5 > ; even NeXTSTEP/MacOS X although based on mach/UNIXaB > > are to a certain degree okay. But I won't accept non-technical! > > arguments for technical ones.. > G > But you're more than willing to offer them.  Most of your "technical" L > arguments are nothing more than emotional crap that have no basis in fact.  ? Unfortunately the opposite is true. And again unfortunately youi: obviously ran out of arguments. You even didn't jump on my: offers to do some real world testing. Why do you continue? You've lost your credibility.s  R > >> Ok, fair enough, but its far too easy for those applauses to turn into thrown# > >> fruit if they do a shitty job.i > >r( > > Not sure I understood you correctly. > M > If VMS engineering messes up the COE implementation then people who are form6 > it now will be against it due to their incompetence.  @ I posted my opionen already there is no chance to change even if> the implementation would not completely destroy VMS' technical superiority.   > > I didn't see a lot peopleoD > > thinking about the risks. I seems that most people here have theA > > attitude that VMS is at risk and any measure could help it. Ir> > > simply share their opinion. Instead I think this is a very@ > > dangerous attitude as in case of danger outmost care must beB > > taken to make the right decisions because it could be the last	 > > ones.h > K > You've never really pointed out concrete risks.  Sure, they might mess upiK > the kernel, but then, they might mess it up when they're adding a non-COE J > feature as well, so that's not really a risk.  "architectural purity" is > is not a risk, either.  ; I pointed out several risks in cov. I got several very gooda7 comments and additional analysis which came to the same ? conclusion that the risk is very high. It seems that you aren'tt8 using VMS enough to understand the techical implications@ discussed so far (to drop only two: directory modification date,< write permission in dir implies delete on file). If you love@ UNIX crap, do so, but leave us alone! We don't need another UNIX@ crap!! Do you understand this? What is your interest in spoiling@ VMS with UNIX crap? Aren't there enough UNIX crap flavour around to play with for you??  Q > >> No arguments there.  If they compromise, I suspect they'll compromise on theo > >> side of VMS.s > >e > >:-( > - > Why is this sad?  FOR VMS, not against VMS.t  8 Because UNIX crap won't do any good for VMS - see above!  O > >> >> You implied that things chosen by the majority are corrupt and or lowerd > >> >> quality. > >> > > >> > No - instead I implied that majority decisions don't warrant quality. And I admit I normally expect that the majority is wrong (Hollywood movies, Windoze, Macdonalds, cars). But basically I prefer analysis - not marketing arguments.  > >> > > >>O > >> Same here...just because the overall trend does tend to show that the moreoB > >> quantity is the lower quality, there's no direct correlation. > > ? > > Statistically there *is* correlation. But correlation is nocC > > proof technically spoken. But you are missing my point that youm@ > > dont' argue logically. I only claimed that the million fliesB > > argument isnt' worth anything. The majority could be right but > > in most cases it is wrong.; > > Remember, you started this quality vs quantity thought.  > 7 > Actually, as I recall we were discussing marketshare.e  : No, we are discussing if UNIX crap on VMS will do any harm6 technically to VMS. So far the case is clear: it will.  
 > > [SNIP] > >lK > >> >> That's odd, because I find unix far simpler to understand than VMS.  > >> >2> >> > As I pointed out: people are different. But besides that the design criticism is beyond liking or disliking. In fact this is an explanation why UNIX sucks so much. What is difficult with VMS to understand? What about the CLI? Do you like shells better than DCL? Do you think man is better than HELP? > >>T > >> No, Help is truly wonderful, but it doesn't allow me to search it for keywords,R > >> but man does.  There are pros and cons to each (though, universally, GNU info > >> sucks). > >a@ > > The search is very simplistic and not very helpful althoug ID > > admit without search is would even more a pain in the arse. OTOHA > > with VMS HELP there is rarely a need to search for a keyword.l > M > There you are wrong.  There are many times where it would be nice to searchfG > help for a word when an error message comes up, or you don't know theo% > commands which might be applicable.c  > There is a different facility for error message searches (with@ explanation) - don't you know that? This is better than any UNIX? crap equivalent. Where would you search for an error message on 
 UNIX crap?> And your example sucks. If you don't know which commands might= be applicable how would you search within man?? Anyway do you  about HELP HINTS?a  B > > And the quality of HELP (not only its structured presentation)C > > is much better than most man pages. Think only about the number\ > > of examples. > I > Sure, it is quite good. No arguments there.  Better than man?  Maybe in E > many cases.  The structured approach is nice, but sometimes it getsaD > in the way, especially if I want to print out the help for a whole
 > command.  = You might have a point here. But honestly I never print out ah9 complete command reference for one command and I wouldn'ta9 hesitate a second if I had to choose between HELP and mane4 because there so many advantages of HELP besides its" completeness and example richness.  J > >> The kernel structure of Unix is easier to understand, at least on the > >> conceptual level. > >aD > > This could well be true because there is hardly any concept (not > > only my opinion).N > P > YES ONLY YOUR OPINION.  You really need to learn that just because you believe! > something doesn't make it true.   = I don't believe it - I *know* it, I've used it, I use it, I'm = confronted with the problems every day. Read "The UNIX-Hatersr> Handbook" - I'm sure the authors are the better engineers than; you. At least on of them is a well known EDP specialist and 8 co-author of a very good book about NeXTSTEP. Ask them!!  h > >> >> >> > I also publicly offered to put money into a fund to get a full VMS Opera port. No response. > >> >> >> >^ > >> >> >> > Do you know on what platform Navigator has been developed? Could it be UNIX crap? > >> >> >> > Navigator is crap. I use it to avoid Micro$oft although some people say the IE is the better browser. If there is a decent VMS browser available I'll switch to that.
 > >> >> >>T > >> >> >> So, we come back to the core of the problem:  lack of available software.V > >> >> >> Whether something was built on unix doesn't make it crap, anymore than unix> > >> >> >> as a whole is crap.  Why don't you write a browser?	 > >> >> >iP > >> >> > Good question. Because I have *other* things to do. Do you understand? > >> >>ER > >> >> Of course I do.  However, others have other things to do, and supporting a- > >> >> small-user-base OS isn't one of them.. > >> > > >> > Not a technical point.U > >>T > >> True, its not.  But it is a valid point:  people aren't going to write softwareP > >> for a platform (regardless of its quality) if there aren't very many people > >> there to use it.  > >t) > > Definitely wrong - think about Linux.  > 0 > Ok, lets ask ourselves:  why is Linux popular? > * >         1) it runs on commodity hardware >         2) the price is rightE. >         3) it is source-compatible with unix  ? Point 3 is completely wrong. BTW have you ever tried to compilea? Maxwell, the text processor? Do that and come back and tell us!X< And there is one additional thing wrong with point 3: beeing< compatible with UNIX isn't a point in favour of Linux at all: because the Linux developers and users give a shit on that@ because they never ever think a second about UNIX which in their: opions is a crap, old, shitty and very expensive OS, which: exists in many incompatible flavours. Why would they care?  $ > Guess what, why isn't VMS popular? > , >         1) it runs on proprietary hardware" >         2) its expensive as hell. >         3) it isn't compatible with anything  0 NOT A TECHNICAL POINT. Go ahead and use Windoze!   [SNIP]  J > >> No arguments there.  But since they're optional, and the user will beS > >> installing them, its their call.  But, you're right, it leads to the "anythingiT > >> via telnet sucks"-syndrome that windows users have because they've got a shitty > >> default telnet client.o > >dC > > Good point and well in line of my fears. Burn a manager and youu/ > > probably won't sell him/her anything later.  > K > Of course, VMS is hardly selling at all now.  Its only a possibility, notn? > a certainty, that its going to create the "it sucks" mindset.t   No comment.a   [SNIP]  S > >> >> Its not impossible to see the beauty in two different methods.  Its easy tomQ > >> >> see advantages to both methods, and disadvantages.  But it all boils down 0 > >> >> to the right tool (as I've said before). > >> >>i
 > >> >> Briann > >> > > >> > These general statements don't tell me much. What are the points which attract you to VMS after you mostly switched to UNIX and wrt the fact that your experience is that VMS is harder to understand than UNIX?i > >>T > >> I like the VMS command line, though its probably because of nostalgia.  I wroteO > >> quite a bit of DCL.  Its a quality OS, that in itself makes it attractive.  > >rB > > Quality is a black box. Sad that this really interesting pointB > > which will very much appropriate to cov didn't fit into your's% > > news readers (read "emacs") page.b > N > Emacs wraps the lines just fine, so I do see them, but it tends to wrap themK > in the middle of words and the newsreader bitches that I have submitted a M > post with long lines...netscape not bitching about long lines is netscape'ss > fault, not Unix's. >  > Briant   Fixed that.u   ------------------------------  % Date: Wed, 21 Mar 2001 10:11:53 +0000N) From: Christof Brass <brass@infopuls.com>hY Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMSEducationalProgram) Prob, Message-ID: <3AB87E69.E14C6F89@infopuls.com>   Bill Gunshannon wrote: > / > In article <87u24o4arv.fsf@prep.synonet.com>,f1 >  Paul Repacholi <prep@prep.synonet.com> writes:  > |>K > |> I have said it before, and I will say it again. Unix is still possiblehJ > |> the system for developing how to solve a problem, or finding out if aA > |> usefull sollution is possible or practicle. It is a very badyI > |> environment to develop a solid code base that can be ported to others > |> platforms.  > E > Are you saying it is trivial to port from VMS to any other system??mE > Surely you jest!!  What other system in the world has LIB$FIND_FILEl > or LIB$DAY or LIB$GETSYI??  ? No he isn't. But VMS isn't lying about that in pretending to bes@ a platform for multiplatform development. VMS is just VMS and it
 is excellent.n  G > It doesn't matter which system your on.  Either you use a minimal setaI > of common calls or you write non-portable programs.  One decides wethernE > or not they are interested in performance, elegance or portability.e' > You can't have all three in any case.e  > Why is this situation then any better than VMS? Obviously from@ your wording the portable UNIX crap apps are not elegant and are# not performing. Well taken. Thanks.e  F > Some programmers go to a lot of effort isolating these machine or OSD > dependancies.  Some do not.  It all depends on the target goals ofF > the program.  Considering how few people have actually heard of VMS,D > much less actually used it, is it any surprise that portability to0 > VMS is so low on the table of considerations??    Where is the technical analysis?   > bill >  > --L > Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesF > bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. > University of Scranton   |@ > Scranton, Pennsylvania   |         #include <std.disclaimer.h>   ------------------------------  % Date: Wed, 21 Mar 2001 15:48:27 +0000 $ From: Steve.Spires@yellowpages.co.ukY Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMSEducationalProgram) Pron/ Message-ID: <00256A16.0056D9DC.00@quegw01.btyp>   L Contact:   Tel: 3063  -  IS - Infrastructure, 1st Floor, Bridge Street Plaza     Bill wrote;w  .  "find . -type f -exec rm -f {} \; ; rm -rf *"  D I'm sorry, I can't find that particular emoticon in any of my books.   ;^De   Steve Spires   [Just a joke Bill!]h        E bill@triangle.cs.uofs.edu (Bill Gunshannon) on 21/03/2001 12:01:30 AMe    To:        Info-VAX@Mvb.Saic.Com+ cc:         (bcc: Steve Spires/YellowPages) P From:      bill@triangle.cs.uofs.edu (Bill Gunshannon), 21 March 2001, 0:01 a.m.  9 Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMSu EducationalProgram) Program)        2 In article <G7W3OjMzxo7TfaW0cHyNnUquOgut@4ax.com>,:  David Beatty <David.Beatty@sasSPAMITUPTHEZZZ.com> writes:D |> On 20 Mar 2001 18:25:35 GMT, bdwheele@indiana.edu (Brian Wheeler)	 |> wrote:u |>M |> >Out of curiosity, is there an easy way to delete a directory structure ont; |> >vms without having to run "DELETE *.*;*" over and over?e |> >	 |> >Brian  |>' |> You probably mean DELETE [...]*.*;*.W |>H |> Depends on how you define easy.  It's not trivial, but can be done inG |> a fairly small (recursive) command procedure -- less than 100 lines,iE |> probably closer to 50 lines.  The F$SEARCH lexical function is the H |> key here.  I've sure that utility has been written by many folks over
 |> the years.a  G And a 50-100 line "small (recursive) command procedure" is simpler than   2      "find . -type f -exec rm -f {} \; ; rm -rf *"  
 in what way??    bill   --J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |> Scranton, Pennsylvania   |         #include <std.disclaimer.h>   ------------------------------  % Date: Wed, 21 Mar 2001 10:22:56 -0500e From: William_Bochnik@acml.comY Subject: RE: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMSEducationalProgram) Proe> Message-ID: <OF81D50DDB.B612A269-ON85256A16.00546A54@acml.com>  A I think the problem was, with the shell expansion and such of thehA filenames, they ended up with an rm command that was too long fort> the shell to handle, which is the source of the problem in the first place.      ]                                                                                              c]                     Christopher                                                              e]                     Smith                        To:  Info-VAX@Mvb.Saic.Com                  Y]                     <csmith@amdoc                cc:                                         i]                     s.com>               Subject:     RE: Dumbing Down VMS with UNIX         v]                                          Elements? (was Re: OpenVMS EducationalProgram)       ]                     03/21/2001           Program) EducationalProgram) Program)               s]                     10:14 AM                                                                  ]                                                                                              u]                                                                                              g       > -----Original Message-----2 > From: Christof Brass [mailto:brass@infopuls.com]  < > But why the hell isn't it possible to simply pipe the find > result into rm?.  9 It is possible to do, depending on your shell, of course,i something like this:I  ! rm -f `find . -name foo\* -print`t  @ This will have the desired affect, I believe.  The simple answer for thehA reason that you can't pipe things into rm is that rm doesn't read  itsg> standard input looking for that kind of thing. :)  Why doesn't it?  I don'tA know.  Could be because they waited until they were about to fill 
 a disk up,$ and threw it together in a hurry. :)   Regards,   Christ  ! Christopher Smith, Perl Developer  Amdocs - Champaign, IL   /usr/bin/perl -e '? print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");  'e          F ______________________________________________________________________  : The information contained in this transmission may contain< privileged and confidential information and is intended only< for the use of the person(s) named above. If you are not the< intended recipient,  or an employee or agent responsible for? delivering this message to the intended recipient,  any review, @ dissemination, distribution or duplication of this communication> is strictly prohibited. If you are not the intended recipient,A please contact the sender immediately by reply e-mail and destroyf# all copies of the original message.n   ------------------------------    Date: 21 Mar 2001 10:37:48 -0500/ From: jordan@lisa.gemair.com (Jordan Henderson) Y Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMSEducationalProgram) Pror* Message-ID: <99ahsc$8rr$1@lisa.gemair.com>  > In article <OF81D50DDB.B612A269-ON85256A16.00546A54@acml.com>,"  <William_Bochnik@acml.com> wrote: >uB >I think the problem was, with the shell expansion and such of theB >filenames, they ended up with an rm command that was too long for? >the shell to handle, which is the source of the problem in thet
 >first place.n >a  F I seem to be the only one pointing out that xargs is perfect for this.  : It also allows you to have scripts that generate groups of parameters to execute.  	 man xargs    >o >e^ >                                                                                             ^ >                    Christopher                                                              ^ >                    Smith                        To:  Info-VAX@Mvb.Saic.Com                  ^ >                    <csmith@amdoc                cc:                                         ^ >                    s.com>               Subject:     RE: Dumbing Down VMS with UNIX         ^ >                                         Elements? (was Re: OpenVMS EducationalProgram)      ^ >                    03/21/2001           Program) EducationalProgram) Program)               ^ >                    10:14 AM                                                                 ^ >                                                                                             ^ >                                                                                              >( >  >  >> -----Original Message-----a3 >> From: Christof Brass [mailto:brass@infopuls.com]o >M= >> But why the hell isn't it possible to simply pipe the find  >> result into rm? >d: >It is possible to do, depending on your shell, of course, >something liker >this: >n" >rm -f `find . -name foo\* -print` >nA >This will have the desired affect, I believe.  The simple answere >for theB >reason that you can't pipe things into rm is that rm doesn't read >its? >standard input looking for that kind of thing. :)  Why doesn'tt
 >it?  I don'ttB >know.  Could be because they waited until they were about to fill >a disk up, % >and threw it together in a hurry. :)a >n	 >Regards,l >o >Chris > " >Christopher Smith, Perl Developer >Amdocs - Champaign, IL  >m >/usr/bin/perl -e 'i@ >print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n"); >' >y >n >T >m >rG >______________________________________________________________________  >r; >The information contained in this transmission may containr= >privileged and confidential information and is intended only = >for the use of the person(s) named above. If you are not ther= >intended recipient,  or an employee or agent responsible fors@ >delivering this message to the intended recipient,  any review,A >dissemination, distribution or duplication of this communicationa? >is strictly prohibited. If you are not the intended recipient,sB >please contact the sender immediately by reply e-mail and destroy$ >all copies of the original message. >o   -Jordan Hendersont jordan@greenapple.coma   ------------------------------  % Date: Wed, 21 Mar 2001 11:17:06 -0500n+ From: Brendan Welch <brendan_welch@uml.edu>bY Subject: Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMSEducationalProgram) Pro & Message-ID: <3AB8D402.ABD2255@uml.edu>   Christopher Smith wrote: >  > > -----Original Message-----4 > > From: Christof Brass [mailto:brass@infopuls.com] > > > > But why the hell isn't it possible to simply pipe the find > > result into rm?  > J > It is possible to do, depending on your shell, of course, something like > this:e > # > rm -f `find . -name foo\* -print`w > G Pardon my getting off into Unix-ism's; I find my experience as detailed , below to be useful (like VMS; dislike Unix).  H I tried the command above on my Unix box (saying to myself, Why didn't IH think of that?).  I replaced -f  with -i purposely to stop as soon as itD found the first file.  I experimented by first performing just the   find . -name foo\ -printH so I would know what I should be finding (was also trying to shorten the9 find command a little bit by leaving off some arguments).n  B This seemed to work as expected when I was far down in a directoryG tree.  The find command rapidly found what I wanted, and worked OK whenc' I embedded it inside the rm -i command.g  F But if I use it in a way like I would in real practice, sitting at theD head of a file system, and looking for a huge bunch of ancient filesH which I want to delete, the command performs badly in 2 ways.  First, itF runs through the find command for the entire disk, before it tries theC first rm; this takes a lot of time.  Secondly, if it finds too manysD files of the desired name, it fails entirely, with the error messageE "arg list too long".  (I think the magic number is 1000, and I do note) know how to change that default on Unix.)     J > This will have the desired affect, I believe.  The simple answer for theG > reason that you can't pipe things into rm is that rm doesn't read itseM > standard input looking for that kind of thing. :)  Why doesn't it?  I don'teN > know.  Could be because they waited until they were about to fill a disk up,& > and threw it together in a hurry. :)  C This brings me to another thing, which I first used on VMS and haveo carried to Unix.E When a disk is filling up and/or a system is creaking to a near-halt,fD and you can hardly execute a command even though you are SYSTEM withC priority 16, that is no time to start searching for files.  You may C disagree with me, but I have saved my backside a couple of times byiF having occasionally run a command which does a directory of the entireC disk, saving the output in a file named allfiles.txt .  If you just G _know_ a file of a certain name exists (or only know part of the name),sB it is a lot quicker to search through allfiles.txt with grep or anH editor, to find the exact location and name, than to do a "find" or "dir0 [...]" in real time at the instant of emergency.     --E Brendan Welch, system analyst, Univ. of Massachusetts - Lowell, W1LPG>   ------------------------------    Date: 21 Mar 2001 13:18:40 +0100G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>t2 Subject: Re: ElSA Gloria Synergy-8 patch confusionH Message-ID: <y466h39uxr.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  3 "Mark D. Jilson" <jilly@clarityconnect.com> writes:y  + > Noted and passed on to the proper people.   = It seems VMS engineering needs a dependency checker for ECOs.v  B I have also noted that the header of the READMEs, which seem to beA automatically generated, either sometimes have things unspecifiedoA (the INSTALL_RATING is most often hit) or show evidence of manualr
 editing...   	Jan   ------------------------------  % Date: Wed, 21 Mar 2001 11:29:35 -0500   From: norm.raphael@jamesbury.com2 Subject: Re: ELSA Gloria Synergy-8 patch confusion4 Message-ID: <C2256A16.0059CB8A.00@jklh21.valmet.com>  . Those doing the READMEs are doing yeoman work.  D That said, sometimes just re-reading the text would be a great help.  O This specific problem this time, however, is that a dependency was created thatl hasyJ no basis.  ECO 2 has a requirement for ECO 1 where ECO 1 is not needed forJ ECO 2 to function correctly.  So an unneeded, do-nothing-here ECO 1 has toO be installed to get the functionality of ECO 2.  No dependency checker is going 
 to find that.>          I jan@mailhost.neuroinformatik.ruhr-uni-bochum.de on 03/21/2001 07:18:40 AMn   To:   Info-VAX@mvb.saic.como cc:    Info-VAX@mvb.saic.com3 Subject:  Re: ELSA Gloria Synergy-8 patch confusions        3 "Mark D. Jilson" <jilly@clarityconnect.com> writes:   + > Noted and passed on to the proper people.   = It seems VMS engineering needs a dependency checker for ECOs./  B I have also noted that the header of the READMEs, which seem to beA automatically generated, either sometimes have things unspecifiedcA (the INSTALL_RATING is most often hit) or show evidence of manual 
 editing...        Jan   ------------------------------   Date: 20 Mar 2001 20:06:58 CDT= From: wayne@tachysoft.xxx.412538.killspam.015a (Wayne Sewell)d$ Subject: Re: Free The OpenVMS Seven!. Message-ID: <3aUCCl4g3pAX@tachxxsoftxxconsult>  _ In article <5.0.2.1.0.20010319122529.01977da0@24.8.96.48>, Dan Sugalski <dan@sidhe.org> writes:S2 > At 05:12 PM 3/19/2001 +0000, David Mathog wrote:9 >>In article <FreeThe.OpenVMS_7@nowhere.nil>, Doc.Cypher a- >><Use-Author-Address-Header@[127.1]> writes:e >> >(b) Business Ready.a >> >9 >> >* Cost around 2^11 USD (**INCLUDING** User Licenses).t >>                ^^^^ >>F >>I sure hope that would include user licenses - that's 200 Billion US
 >>dollars. > : > That looks like 2^11, not 2*10^11. 2^11 is only $1024...    L You could probably still buy a Congressman for that, though.  Maybe a little more for a Senator.  :-)     -- dO ===============================================================================uM Wayne Sewell, Tachyon Software Consulting  (281)812-0738  wayne@tachysoft.xxxn: http://www.tachysoft.xxx/www/tachyon.html and wayne.html  K change .xxx to .com in addresses above, assuming you are not a spambot  :-)NO ===============================================================================UO Dean Wormer to Flounder: "Fat, drunk, and stupid is no way to go through life."C   ------------------------------  # Date: Wed, 21 Mar 2001 16:47:56 GMTv/ From: mbecker@americusglobal.com (Marie Becker)@G Subject: Help ................... Need Sybase DBA with VMS skills    !!s. Message-ID: <3ab8db25.76964045@news.dscga.com>  C Our client in New York City is currently looking for an experiencedu- Sybase Administrator with DEC VAX/VMS skills.s    6 This position is in the New York financial district.    : If available, please respond with a copy of your resume to      
 Roger Shankert Account Managero   Americus Global Software, Inc. Atlanta, GA    rshanker@americusglobal.comI   ------------------------------  % Date: Wed, 21 Mar 2001 11:02:40 -0500i6 From: "Dominic Olivastro" <DOlivastro@ChiResearch.com>" Subject: HELP on License Requestor% Message-ID: <T94u6.16$3i3.301@client>t  L I keep posting questions about PW and Win 2K but no one answers.  Is there a strike?i  K I am having trouble getting  the license requestor to find licenses from my K Windows 2000 client.  has anyone gotten this working?  Is there some trick.eF When I start the PC, I get the usual box from Lan Manager saying it isL searching for the license server.  This goes on for some time and then I get+ a message saying the license was not found.g  	 My Setup:  Windows 2000
 PW Client 7.2e   VAX VMSe PW Server v5.0F    TIA  DOMl     -- Dominic Olivastrol CHI Research, Inc. 10 White Horse Pike  Haddon Heights, NJ 08035   Phone:  1-856-546-0600 Fax:       1-856-546-9633p% mailto:    DOlivastro@ChiResearch.como   ------------------------------    Date: 22 Mar 2001 02:15:24 +0800, From: Paul Repacholi <prep@prep.synonet.com>( Subject: Info on low level driver access- Message-ID: <87g0g77zur.fsf@prep.synonet.com>   @ I've gone through the driver docs, and the ARM, and even bits of/ the internals manual, but can't find an answer.d  F I want to get the interupt from the on chip console port. It is listedC in the ARM, but I can not find any info on how to be sure that somemF other code, like a kernel debugger, has already set it up. Where can I find this info?0   -- .< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.o@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  % Date: Wed, 21 Mar 2001 13:38:17 -0500t% From: "Brian Tillman" <tillman_brian>n' Subject: Re: Internet Printing Protocolf$ Message-ID: <3ab8f560$1@news.si.com>  J >Any idea if OpenVMS or any of the related printer softwares  will support >the MS + HP >technology ???   K Process Software is in the midst of adding it to Multinet (and, presumably,m	 TCPware).  --A Brian Tillman                   Internet: tillman_brian at si.comtA Smiths Aerospace                          tillman at swdev.si.com8= 3290 Patterson Ave. SE, MS      Addresses modified to prevent>< Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------  % Date: Wed, 21 Mar 2001 11:52:56 -0700.% From: Dan O'Reilly <dano@process.com>e' Subject: Re: Internet Printing ProtocoldA Message-ID: <5.0.2.1.2.20010321115233.00a97918@ntbsod.psccos.com>n  + At 11:38 AM 3/21/2001, Brian Tillman wrote:TL > >Any idea if OpenVMS or any of the related printer softwares  will support > >the MS + HP > >technology ???> >eL >Process Software is in the midst of adding it to Multinet (and, presumably,
 >TCPware).  E You presume correctly.  It's in beta for both (or will be very soon).l     ------I +-------------------------------+---------------------------------------+lI | Dan O'Reilly                  |                                       |yI | Principal Engineer            |  "Why should I care about posterity?  |eI | Process Software              |   What's posterity ever done for me?" |0I | http://www.process.com        |                    -- Groucho Marx    |II +-------------------------------+---------------------------------------+D   ------------------------------  # Date: Wed, 21 Mar 2001 13:51:04 GMT - From: "Dave Pampreen" <davepampreen@home.com>,( Subject: Is there extra logging in BIND?= Message-ID: <cl2u6.83334$W05.16435095@news1.rdc1.mi.home.com>   J I have had a request to get extra info from BIND.  Is it possible to get aF log of the requests, the ip address of the machine requesting, and the* ip/node being translated.  Something like:  D node MYPC.DOMAIN.COM requests WWW.SOMEWHEREONTHEWEB.COM 192.12.34.56  0 I'm running VMS 7.2-1 with TCP/IP Services V5.0A  2 I've looked in the DOC's and haven't foudn it yet.   Dave   ------------------------------  % Date: Wed, 21 Mar 2001 08:34:55 +0100 , From: Nazim MANSER <Nazim.Manser@socgen.com>0 Subject: Re: login failure when sysuaf is lockedT Message-ID: <04CAD3AB8599F014*/c=FR/admd=ATLAS/prmd=SG/o=INFI/s=MANSER/g=NAZIM/@MHS>   I have tryed  the followingo   from terminal 1:  
 $mc authorize  UAF>     from terminal 2M  
 VAX6> set h 0       ?                         Welcome to VAX/VMS V6.2     on VAX6            Username: MANSER
 Password:       0         Welcome to VAX/VMS V6.2     on node VAX6  :     Last interactive login on Wednesday, 21-MAR-2001 08:33>     Last non-interactive login on Wednesday, 21-MAR-2001 08:30  +             You have 339 new Mail messages.     0  we are running a cluster vms 6.2, vax4000-705A.   reguards, nazim Manser   ------------------------------  % Date: Wed, 21 Mar 2001 09:41:59 +0100h, From: Nazim MANSER <Nazim.Manser@socgen.com>0 Subject: Re: login failure when sysuaf is lockedT Message-ID: <065183AB869570D2*/c=FR/admd=ATLAS/prmd=SG/o=INFI/s=MANSER/g=NAZIM/@MHS>  "  hoffman@xdelta.zko.dec.com wrote:  E >  I've seen some weird stuff happen with queues when multiple SYSUAFa  > files are used.  B >  Given the PID shown in the accounting record, this is clearly a  8 >  cluster.  Please issue the following SYSMAN commands:  + >  Display the settings across the cluster:   > SYSMAN> SET ENV/CLUSTER  K >  check each logical name displayed, ensuring the translation is the same:_!  > SYSMAN> DO SHOW LOGICAL SYSUAF_  < >  check each FID displayed, ensuring the value is the same:#  > SYSMAN> DO DIRECTORY/FILE sysuafa   SYSMAN> do dir/sec/file sysuaf  0 %SYSMAN-I-OUTPUT, command execution on node VAX6 Directory DSA4:[CLUSTER_COMMON]oZ SYSUAF.DAT;1         (964,50378,0)        [SYSTEM]                         (RWED,RWED,RE,) Total of 1 file.0 %SYSMAN-I-OUTPUT, command execution on node VAX7 Directory DSA4:[CLUSTER_COMMON]dZ SYSUAF.DAT;1         (964,50378,0)        [SYSTEM]                         (RWED,RWED,RE,)   the multiple batch jobs runs under the same username (production) but the logfile LISTE_ICW400.LOG is not created. (normal because the batch login failed) we are using the same sysuaf.D| maybe the login failed because the record (noninteractive login time) was locked by the same username during another submit.. as workaround i put a $wait before the submit.P the ob runs every day and during this month failed only 2 times because of this.    
 reguards..   ------------------------------    Date: 21 Mar 2001 12:57:24 +0100G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>o  Subject: Re: MOUNT/BIND too slowH Message-ID: <y4bsqv9vx7.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  M I don't think using RAID-5 for a write-heavy disk is a good idea, unless yourt$ disk space is extremely constrained.   	jan   ------------------------------  % Date: Wed, 21 Mar 2001 09:44:33 -0500 ' From: Jack Bogart <jbogart@vangard.com>  Subject: NAS/SAN6 Message-ID: <002201c0b215$72780720$37010c0a@akron.com>  L Is anyone mounting OpenVMS volumes to a NAS (Network Appliance Filer), or onK a SAN (other than EMC or StorageWorks)?  Any performance issues?  Problems? 	 Warnings?     < ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Jack L. Bogart Solutions EngineerJ StorNet, Inc.          175 Montrose West Avenue; Suite 280          Akron, Ohio   44321D Office (330) 666-5190     Fax (330) 666-0830     Cell (330) 284-7471H Centralized Storage Management --- High Availability --- Data ProtectionE Outsourced Storage Operations --- Consulting --- Support --- Training J "I feel that there is a world market for as many as five computers" Thomas Watson, IBM corp. - 1943< ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~   ------------------------------  % Date: Wed, 21 Mar 2001 16:20:26 +0000 0 From: andrew harrison <andrew.nospam@uk.sun.com>( Subject: Re: OpenVMS Educational Program* Message-ID: <3AB8D4CA.61A58036@uk.sun.com>   Christof Brass wrote:r >  > andrew harrison wrote: > >r > > Christof Brass wrote:. > > >f[ > > > Great! I'm on the right way. Having Andrew as opponent is as safe as sleeping in bed.e > > >o > >t7 > > Your bed must be a very very scary place. And don'ti6 > > get out in the dark the monster inder the bed will > > get you. > > < > > Strangely you are only on the right track if your secret( > > agenda is the extinction of OpenVMS. > ? > The opposite is the case and maybe you know it. It is like inhB > some stories where the bad people behave as if they want to help > but instead the do the worst.^ > 9 > > Even stranger I disagree with you, not because I carer5 > > about the survival of OpenVMS but because you are_
 > > wrong. >  > Fits in the picture. > ; > > You have been quite happy to trash COE without offeringf7 > > any suggestions as to how you would get major ISV'so: > > to support OpenVMS. Its terribly negative and probably9 > > rather dispiriting to people like Fred who appears tov% > > be involved with the COE process.p > = > Boring. I and others suggested several viable solutions nott< > having the mentioned and analysed risks. Unfortunately youB > didn't get them or you didn't understand them. Bad luck for you.? > Don't use other people for supporting your Anti-VMS attitude. B > And another point: you didn't put in any valuable argument. Your > post is an empty suit.  = Really so what were they ? perhaps a recap would be helpfull.l  = But in this case the only person showing an Anti-VMS attitude:; is you, you may not think so. But do you really think that 57 repeating sections of the UNIX haters handbook, an old i8 and even at the time not very accurate set of anti UNIX ) opinion's is going to advance your case. t  F All you are doing is providing all the ammunition that any UNIX admin/A ISV/advocate would need if they wanted to make a case that people > who use OpenVMS are either criminally uniformed or just plain > bizzare. With your kind of views you will get people sticking A red tape and "Danger OpenVMS do not enter" signs up around their  
 AlphaServers.x  - This is hardly helpfull to OpenVMS nor is it a advancing your case either.   ; > > So how would you get major ISV's to support OpenVMS ???m >  > See above!  - No you havn't answered the question, put yourI' points into short lines and send them.     > ; > > And don't say OpenVMS is great and UNIX is crap becasuef> > > even if it was true it isn't going to get ISV's to support > > OpenVMS. > ? > VMS is really great and UNIX is almost as crappy as Windoze -w= > unfortunately in several respects its much more crappy than B > Windoze (I'm sorry to state this) - the most important advantage= > is that because its simplicity you can survive whereas withf > Windoze you sometimes can't. >   6 Christof you havn't answered the question. Why do you 5 think UNIX is crap ? repeating add-nauseum that it ise0 without providing any coeherant examples is not  going to win you the argument.    ? > You obviously didn't get is: COE and UNIX on VMS won't do the  > trick!  8 Why ? you havn't articulated any response to this except8 what appears to be a gut fear that it will somehow break OpenVMS.  5 Do you think that the X-Files are documentary's ?????t   Regards' Andrew Harrisono Enterprise IT Architecta   ------------------------------  % Date: Wed, 21 Mar 2001 10:23:03 +0000?) From: Christof Brass <brass@infopuls.com>w, Subject: Re: POSIX Streams, File Permissions, Message-ID: <3AB88107.E3AB1BB8@infopuls.com>   Hoff Hoffman wrote:  > Z > In article <3AB6C871.D9DEDB24@infopuls.com>, Christof Brass <brass@infopuls.com> writes:9 > :Why can't we have a pure VMS version and UNIX-version?e > F >   Your question is sufficiently terse as to be exceedingly difficult' >   to answer.  In other words, "huh?".  > P >  ---------------------------- #include <rtfaq.h> -----------------------------L >       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.comP >  --------------------------- pure personal opinion ---------------------------N >    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   Sorry.: Can we have a VMS which is free of UNIX API in the kernel?7 Can we have a VMS which behaves as before wrt directorye- modification date and delete file permission?s   ------------------------------  % Date: Wed, 21 Mar 2001 06:45:07 -0500-  From: John Santos <JOHN@egh.com>, Subject: Re: POSIX Streams, File Permissions6 Message-ID: <1010321063234.21646A-100000@Ives.egh.com>  * On Wed, 21 Mar 2001, Christof Brass wrote:   > Hoff Hoffman wrote:l > > \ > > In article <3AB6C871.D9DEDB24@infopuls.com>, Christof Brass <brass@infopuls.com> writes:; > > :Why can't we have a pure VMS version and UNIX-version?e > > H > >   Your question is sufficiently terse as to be exceedingly difficult) > >   to answer.  In other words, "huh?".o > > R > >  ---------------------------- #include <rtfaq.h> -----------------------------N > >       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.comR > >  --------------------------- pure personal opinion ---------------------------P > >    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com >  > Sorry.< > Can we have a VMS which is free of UNIX API in the kernel?9 > Can we have a VMS which behaves as before wrt directory / > modification date and delete file permission?m  D "wrt directory modification date"...  Are you refering to the changeF in behaviour of VMS where a directory's modification date gets changedB if it is extended or truncated, that I reported several weeks ago?  ? This is a bug, not an intentional change.  I don't think it hasr@ anything to do with COE.  (If COE requires that a directory have> a modification date, and that that date get changed whenever a= file is created or deleted in that directory, then this is anAA entirely different issue.  This also is *NOT* what happens in VMS@C with the new behaviour/bug in effect.  If COE has this requirement,eA then the correct solution is to add a new date attribute, call itCB a "revision date" if you prefer, that behaves as COE requires, but< keep the modification date as is.  Or make the revision dateA behave as the modification date used to, make the mod date changecB on each file creation/deletion, and make backup check the revisionB date, not the modification date, for incremental backups.  I don't! care, as long as it works right.)u  @ The reason I say it is a bug is that I discovered it only occursD on systems that have had the Alpha VMS V7.2-1 F11X-V0200 ECO applied
 to them.    D Also, Compaq support has informed me that there is a patch availableD to fix the problem, but at latest attempt, I wasn't able to downloadC it.  (Either they gave me an incorrect filename or ftp server name,R@ or there is some kind of protection problem accessing the file.)& So I haven't been able to test it yet.   -- p John Santosi Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Wed, 21 Mar 2001 12:26:40 -0500o- From: "Peter Weaver" <peter.weaver@stelco.ca>a, Subject: Re: POSIX Streams, File Permissions4 Message-ID: <Mv5u6.163322$Z2.2016792@nnrp1.uunet.ca>  6 "Christof Brass" <brass@infopuls.com> wrote in message& news:3AB88107.E3AB1BB8@infopuls.com... >...< > Can we have a VMS which is free of UNIX API in the kernel?9 > Can we have a VMS which behaves as before wrt directory / > modification date and delete file permission?=  C Christof, I've only been glancing at the other thread(s) so I don'teC know if you have actually read the "OpenVMS Alpha Version 7.2-6C1 -h( New Features and Release Notes" documentF (http://www.openvms.compaq.com/solutions/government/coe/AA-RNALA-TE.PD" F) but in it there is the section;  & -------------------------------------- 3.3.1 Additional File Times3B Access dates reflect the last time a file was accessed. To support POSIX-compliantfF file timestamps on ODS-5 disks, OpenVMS Alpha Version 7.2-6C1 includes threes new file attributes: . ATR$C_ACCDATEr . ATR$C_ATTDATEl . ATR$C_MODDATEoF ATR$C_ACCDATE (corresponds to POSIX st_atime) reflects the last time a@ file was accessed. ATR$C_ATTDATE (corresponds to POSIX st_ctime) reflects: the last time a file attribute was modified. ATR$C_MODDATE (corresponds to0F POSIX st_mtime) reflects the last time data was modified. Since access datescE must be written out to disk, there is a performance impact when theseg new file@ attributes are used. The system manager can use the SET VOLUME / [ENABLEn? /DISABLE]=ACCESS_DATES command to enable or disable access date  support , and the frequency for changing access dates.' ---------------------------------------P  D So if you never do a $ SET VOLUME /ENABLE=ACCESS_DATES then you will$ not have to worry about these dates.  ) Also if you never issue the command $ SETcE PROCESS/CASE_LOOKUP=SENSITIVE then you don't have to worry about case 5 sensitive files. If you never issue the command $ SETl@ VOLUME/ENABLE=HARDLINK then you don't have to worry about havingE hardlinks on your disks. Most of the GID stuff looks like it is usingoA the standard SYSUAF ADD/ID, GRANT/ID etc so I don't think that is  going to affect your systems.8  F Basically I don't think any of the work the VMS group is doing for COE? is going to make any major changes to VMS unless you enable it.    ------------------------------  # Date: Wed, 21 Mar 2001 18:11:01 GMTn2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman), Subject: Re: POSIX Streams, File Permissions. Message-ID: <V86u6.8$fB6.434@news.cpqcorp.net>  Y In article <1010321063234.21646A-100000@Ives.egh.com>, John Santos <JOHN@egh.com> writes:  ..E :"wrt directory modification date"...  Are you refering to the change G :in behaviour of VMS where a directory's modification date gets changed C :if it is extended or truncated, that I reported several weeks ago?t  "   These two items are unrelated.    D   DII COE requires a file last access date.  This is support is not C   related to the error in the directory modification date handling.j  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------    Date: 21 Mar 2001 13:43:50 +0100G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>iV Subject: Re: POSIX Streams, File Permissions (was: Re: The Chris and Bill show. Act IIH Message-ID: <y43dc79trt.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  / koehler@encompasserve.org (Bob Koehler) writes:   L > >   (Pardon my cringe.)  I have no doubt access to the device driver will J > >   be available, I just don't know if it will be officially documented.  L What does the device driver implement, in addition to doing I/O to disk, and  what does the C RTL have to do?   F > Interesting.  Of course, I also wanted C RTL access to keyed indexedJ > files and didn't get that (just to the same level of support Fortran has > in it's I/O statements).  K Tsk, tsk, Bob - you're wanting to have your cake and eat it, too. First you-L want C to do Unix-style files, and now it should do VMS-style files as well!   	Jan  # PS: cum grano salis and all that...a   ------------------------------    Date: 21 Mar 2001 08:43:57 -0500- From: koehler@encompasserve.org (Bob Koehler)yV Subject: Re: POSIX Streams, File Permissions (was: Re: The Chris and Bill show. Act II3 Message-ID: <EsyW1LpYM$3o@eisner.encompasserve.org>-   In article <y43dc79trt.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>, Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> writes:  M > Tsk, tsk, Bob - you're wanting to have your cake and eat it, too. First you5N > want C to do Unix-style files, and now it should do VMS-style files as well!  H I promise only to ask for everything on VMS.  I will never, ever ask for everything any where else.  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporationh= NASA GSFC Flight Software       | Federal Sector, Civil GroupfE                                 | please remove ".aspm" when replying    ------------------------------    Date: 21 Mar 2001 21:42:06 +0800, From: Paul Repacholi <prep@prep.synonet.com>V Subject: Re: POSIX Streams, File Permissions (was: Re: The Chris and Bill show. Act II- Message-ID: <874rwnb5n5.fsf@prep.synonet.com>i  ! To go back to file permissions...r   The POXUX spec requires that:   E if the file protection is set to sy:rw,ow:rewd,gr:,wo: then the owner  can not access the file!  * This is the sort of semantic clash I mean.   -- n< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.-@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------    Date: 21 Mar 2001 13:53:39 +0100G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>e  Subject: Re: RMS and file types.H Message-ID: <y4zoef8er0.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  . Paul Repacholi <prep@prep.synonet.com> writes:  = > RSX treated the record length as a 16 bit unsigned with thet? > majik-happens reserved value of 177777 as 'go to next block',eG > VMS/ODS-2 defines it as a signed value. ( Makes you wonder what a -42e > byte record means... )  I Yes, I have never understood that. As the marker value is odd, it doesn'tTH consume a valid record length. Or, if the RMS designers for VMS had beenK "far-sighted", they might have hacked the record length to be unsgned _and_iI in 2-byte units, which means the max record length would be 128 kBytes-2.aH Or they would have reserved space at least for a file format with 32-bit record lengths...e  M Hoff, is there anyone who knows why RMS/VMS chose record lengths signed, when < the compatibility target, RMS/RSX, apparently used unsgined?   	Jan   ------------------------------    Date: 21 Mar 2001 21:38:26 +0800, From: Paul Repacholi <prep@prep.synonet.com>  Subject: Re: RMS and file types.- Message-ID: <878zlzb5t9.fsf@prep.synonet.com>d  I Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> writes:a  0 > Paul Repacholi <prep@prep.synonet.com> writes:  ? > > RSX treated the record length as a 16 bit unsigned with the A > > majik-happens reserved value of 177777 as 'go to next block',aE > > VMS/ODS-2 defines it as a signed value. ( Makes you wonder what ar > > -42 byte record means... )  C > Yes, I have never understood that. As the marker value is odd, it-E > doesn't consume a valid record length. Or, if the RMS designers foruF > VMS had been "far-sighted", they might have hacked the record lengthA > to be unsgned _and_ in 2-byte units, which means the max recordmE > length would be 128 kBytes-2.  Or they would have reserved space aty7 > least for a file format with 32-bit record lengths...l  D This would mean you could never have an odd length record :( The wayD it is now, it trims the max record size by one byte. I can live with that trade off!   B > Hoff, is there anyone who knows why RMS/VMS chose record lengthsA > signed, when the compatibility target, RMS/RSX, apparently used  > unsgined?a   -- r< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.~@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------    Date: 21 Mar 2001 16:21:39 +0100G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>i  Subject: Re: RMS and file types.H Message-ID: <y4elvr87wc.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  . Paul Repacholi <prep@prep.synonet.com> writes:  F > This would mean you could never have an odd length record :( The way7 > it is now, it trims the max record size by one byte. -   E Ah yes, something wasn't quite engaged in my brain when I wrote that..J The record as written is rounded up to an even number (keeping alignment),H but the length specified and returned to the user can be odd, of course.   	Jan   ------------------------------    Date: 22 Mar 2001 01:51:56 +0800, From: Paul Repacholi <prep@prep.synonet.com>  Subject: Re: RMS and file types.- Message-ID: <87snk780xv.fsf@prep.synonet.com>e  I Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> writes:   0 > Paul Repacholi <prep@prep.synonet.com> writes: > H > > This would mean you could never have an odd length record :( The way9 > > it is now, it trims the max record size by one byte. o >  oG > Ah yes, something wasn't quite engaged in my brain when I wrote that.hL > The record as written is rounded up to an even number (keeping alignment),J > but the length specified and returned to the user can be odd, of course.  B Yep, and the extra zero octet is droped in the bin. This can cause@ people to wonder why 'part of the file id missing' if they don't understand the principles.   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.e@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  % Date: Wed, 21 Mar 2001 09:07:59 -0500t# From: Jim Agnew <agnew@hsc.vcu.edu>t7 Subject: Re: Senior OpenVMS systems manager - SINGAPORE + Message-ID: <3AB8B5BF.8AD0D1F4@hsc.vcu.edu>   . Too sensible, they will never fall for that...   William Hymen wrote: > 4 > I really get a kick out of these ads. Is there anyF > job out there for someone less than a " Senior System Administrator"- > Do "seniors" just ripen and fall off trees?o > : > Give me a break, spend a few bucks on training on a less4 > senior guy; you will get your money back 100 times > in dedication. >  > Bill > 3 > <Steve.Spires@yellowpages.co.uk> wrote in messageU+ > news:00256A14.005129EE.00@quegw01.btyp...a > > cc:a > > bcc:J > > Contact:   Tel: 3063  -  IS - Infrastructure, 1st Floor, Bridge Street > Plaza. > > . > > Senior OpenVMS systems manager - SINGAPORE > >  > >aM > > I've had permission from the guy who mail this to me [Craig Humphreys] tos > passK > > this on as I can't, unfortunately, take the second interview because myw
 > wife hasI > > changed her mind [is that grounds for divorce? ;^D ] so if anyone outn > theree. > > would fancy this, get in touch with Craig. > >dM > > By the way, having been there before, Singapore is a terrific environment  > to > > work in. > >i > > Steve Spires > > E > >   ---------------------- Forwarded by Steve Spires/YellowPages ona > 19/03/2001( > > 02:43 PM --------------------------- > >t > >hG > > Steve Spires <stevespires@compuserve.com> on 19/03/2001 09:18:13 AMa > >v' > > To:        Steve Spires/YellowPagesy > > cc:rM > > From:      Steve Spires <stevespires@compuserve.com>, 19 March 2001, 9:18n > a.m. > >e. > > Senior OpenVMS systems manager - SINGAPORE > >d > >e > >a > >i > >i > > 3 > > -------------Forwarded Message-----------------n > >tC > > From:     "Craig Humphreys", INTERNET:craig@eurotechnique.co.ukw- > > To:  , INTERNET:craig@eurotechnique.co.ukc > >d > > Date:     08/03/01 10:50 > >a3 > > RE:  Senior OpenVMS systems manager - SINGAPOREp > >. > >oI > > Please pardon the intrusion, I am currently seeking resources for therF > > following Senior OpenVMS systems manager - SINGAPORE - Permanent /
 > Salaried > > PositionJ > > If interested please call me for further info (PS I will not be in the' > > office tomorrow - Friday 9th March)B > >.; > > *** Family Relocation package for 2 year commitment ***c > >AH > > To support Business Critical Services customers e.g. banks,telecoms,) > > including proactive/reactive servicesn > >X > > Essential:D > > Min 5 yrs (ideal>10) OpenVMS systems management V6.2 to V7.2-1H1@ > > * VAX & Alphas from AS1000 thru to GS140 or better GS160/320= > > *managing large OpenVMS clusters in business critical env ! > > *DECnet Phase IV & DECnet/OSIt > > *TCP/IP Digital UCXn > >o > > Desirable: > > *Crashdump analysisp$ > > *internals programming/knowledge > > *X25 SNA > > *Digital Unix/Compaq TRU64 > >nE > > ***** PLEASE NOTE WE HAVE NOW MOVED TO LONDON - NEW DETAILS BELOWc > *********U > >P > > Kind Regards,a > >h > > Craig Humphreyse > > Managing Directorr > >oJ > > EuroTechnique Consulting - Specialists in IT Recruitment & ConsultancyG > > 61 The London Fruit & Wool Exchange, Old Spitalfields Fruit Market,.$ > > Brushfield Street, London E1 6EX$ > > Direct Line: +44 (0)20 7422 0909$ > > Switchboard: +44 (0)20 7422 0900 > > Mobile: +44 (0)7970 212909 > > Fax: +44 (0)20 7247 9154$ > > Email: craig@eurotechnique.co.uk$ > > Website: www.eurotechnique.co.uk > >iJ > > This Email and any attachments should be read only by those persons to > whomL > > they are addressed. The information or views expressed in this Email areM > > those of the individual sender and not Euro-Technique (Computer Services)-H > > Ltd. (EuroTechnique Consulting). EuroTechnique Consulting accepts noM > > responsibility for any loss or damage incurred through use of this Email.-A > > Any form of reproduction, dissemination, copying, disclosure,  > modification,nI > > distribution and/or publication of this email is strictly prohibited.  > >  > >v > >M > >c > >f > >fL > > ----------------------- Internet Header --------------------------------% > > Sender: craig@eurotechnique.co.ukfD > > Received: from btclick.com (mta02.btfusion.com [62.172.195.247])G > >      by sphmgaab.compuserve.com (8.9.3/8.9.3/SUN-1.9) with ESMTP idd > FAA26262;g- > >      Thu, 8 Mar 2001 05:50:33 -0500 (EST)r: > > Received: from WinProxy.anywhere ([213.123.184.24]) byG > >           btclick.com (Netscape Messaging Server 4.05) with SMTP ids: > >           G9VKS702.1DP; Thu, 8 Mar 2001 10:50:31 +0000K > > Received: from 172.20.1.103 by 172.20.1.202 (WinProxy); Thu, 8 Mar 2001n
 > 10:52:58	 > > +0000d7 > > From: "Craig Humphreys" <craig@eurotechnique.co.uk> # > > To: <craig@eurotechnique.co.uk> 7 > > Subject: Senior OpenVMS systems manager - SINGAPORE ( > > Date: Thu, 8 Mar 2001 10:51:52 -00009 > > Message-ID: <001401c0a7bd$c90b4370$670114ac@etws0003>e > > MIME-Version: 1.0c > > Content-Type: text/plain;r > >      charset="iso-8859-1"i# > > Content-Transfer-Encoding: 7bita > > X-Priority: 3 (Normal) > > X-MSMail-Priority: Normal @ > > X-Mailer: Microsoft Outlook CWS, Build 9.0.2416 (9.0.2910.0) > > Importance: Normal< > > X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4522.1200 > >  > >I > >o > >   > > [Information] -- PostMaster:H > > This transmission is intended solely for the addressee(s) and may beK > > confidential. If you are not the named addressee, or if the message hasl > beenF > > addressed to you in error, you must not read, disclose, reproduce, > distribute orn > > use this transmission. > >eL > > Delivery of this message to any person other than the named addressee is > noteL > > intended in any way to waive confidentiality.  If you have received thisJ > > transmission in error please contact the sender or delete the message. > >o > > Thank you. > >" > >e   ------------------------------  # Date: Wed, 21 Mar 2001 18:41:17 GMTg2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)@ Subject: Re: STARLETSD and SDL/NOPARSE (was: RE: sytem logicals)/ Message-ID: <hB6u6.12$fB6.354@news.cpqcorp.net>   ] In article <CIEJLCMNHNNDLLOOGNJIMEMECDAA.tom@kednos.com>, Tom Linden <tom@kednos.com> writes:e@ :Thanks for the response, what I was looking to determine was ifM :SYS$LIBRARY:STARLETSD.TLB existed on all later VMS systems.  Consensus seemse; :so.  Curiosity caused me to ask the more general question.t  F   STARLETSD is completely unrelated to discussions of system logicals.  K   I will assume this is related to the use of SDL/NOPARSE for the creation -K   of SDL-based definition files for a particular language.  If so, the use eM   of SDL/NOPARSE and STARLETSD mechanism has been used for many years within aL   many language kits.  SDL/NOPARSE was somewhat more difficult to use prior G   to the availability of SDL itself on the Freeware, but it was clearlyCC   possible for those languages that had the necessary SDL backends.w  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------    Date: 21 Mar 2001 14:08:39 +0100G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>n# Subject: Re: Support of old systemsaH Message-ID: <y4wv9j8e20.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  4 hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) writes:  E >   Decade-old computers are very slow and very constrained, and are   >   more expensive to support.  F While true, some have widgets that aren't available for newer systems,C and as long as they manage to get the work done, there is no strongoG reason to replace them. Have a look at the real-time system controllingkF the space shuttle - the design was old when it first flew in 1981, andF the hardware must be over ten years old now: they replaced them with aF new version about that time, and two of the things necessiated by the H upgrade were writing a task that does memory scrubbing, and changing theH "freeze-dried" state - the initial version used core, which is much lessH susceptible to single-event upset (aka bit flips) and retains state whenL powered off. And they are still doing active software development on them...J which is likely not the usual case for such hardware, so the question doesC remain whether support of old hardware in new software/OS releases f( (especially major releases) is required.  J >   Tru64 UNIX specifically decided to discontinue support of TURBOchannelK >   systems, and to focus engineering and qualification efforts on PCI and t' >   new widget and new feature support.   K But the old stuff should continue to work, subject to resource constraints.-  L >   Further, some application vendors and some customers are considering an H >   EV56 minimum support level, in order to use tje byte-word and other J >   instruction set extensions not available in hardware in earlier Alpha F >   microprocessors.  OpenVMS Engineering cannot use as many of these I >   instructions in the OpenVMS executive as we would like to at present  F >   (though the compiler teams have found some rather elegant ways to J >   incorporate these instructions into the newer executable code), which J >   means that there is a performance trade-off across the generations of  >   Alpha systems.  H These two reasons are much stronger, IMO, than the others. Are among theK elegant ways little macros that allow the boot strap code to patch the restnH of the exec to use the appropriate instruction sequences - like the "bigH memory" (more than 32k PFNs) support did on the VAX? (One of the neatest hacks I've seen, BTW.)   	Jan   ------------------------------    Date: 21 Mar 2001 21:35:41 +0800, From: Paul Repacholi <prep@prep.synonet.com># Subject: Re: Support of old systemss- Message-ID: <87d7bbb5xu.fsf@prep.synonet.com>u  4 hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) writes:  L >   Further, some application vendors and some customers are considering an H >   EV56 minimum support level, in order to use tje byte-word and other J >   instruction set extensions not available in hardware in earlier Alpha F >   microprocessors.  OpenVMS Engineering cannot use as many of these I >   instructions in the OpenVMS executive as we would like to at present eF >   (though the compiler teams have found some rather elegant ways to J >   incorporate these instructions into the newer executable code), which J >   means that there is a performance trade-off across the generations of  >   Alpha systems.  C Considering that theAlpha designers made out a VERY strong case whyy? you did not want to do byte-word loads or stores because of thetB performance cost, but this was relaxed because of the need for oddE reads and writes into IO space, Why do the byte/word instructions getSF into the act at all? Is there a rewrite of the Alpha design paper that explains this some where?   4 The other stuff, yes, that I can see be ing usefull.  F BTW, is not the PAL code meant to implement all new Alpha instructions on older machines?   -- e< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.4@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------    Date: 21 Mar 2001 16:19:47 +0100G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> # Subject: Re: Support of old systems H Message-ID: <y4hf0n87zg.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  . Paul Repacholi <prep@prep.synonet.com> writes:  H > BTW, is not the PAL code meant to implement all new Alpha instructions > on older machines?  F No, they resulted in an UUO exception 8-) and the OS has to emulate itF then. Much more costly than generating a sequence of ZAPs and so on in place.   	Jan   ------------------------------  % Date: Wed, 21 Mar 2001 10:35:10 -0500c2 From: rdeininger@mindspring.com (Robert Deininger)# Subject: Re: Support of old systemsnL Message-ID: <rdeininger-2103011035110001@user-2ivea7n.dialup.mindspring.com>  < In article <87d7bbb5xu.fsf@prep.synonet.com>, Paul Repacholi <prep@prep.synonet.com> wrote:    H > BTW, is not the PAL code meant to implement all new Alpha instructions > on older machines?  I It's meant to, and in the few cases I have bumped into, it does.  There's H supposed to be PALcode in there to deal with unimplemented opcode traps.J Look at the /OPTIMIZE or /TUNE qualifiers for your favorite DEC compiler. H They typically say that wrongly-tuned code will still execute correctly.  
 However...J PALcode traps are slow.  They typically come with several memory barriers,I pipelines are drained, etc.  I understand not wanting to put up with this-J is frequently-used code in the VMS executive.  It would still work, but it' might be _way_ slower than native code.:  D There are CPU-dependent modules in VMS, and the appropriate ones areI loaded at boot time.  If the code is isolated, the compiler can sort this D out, making as many versions of the code as needed.  But in a deviceJ driver, for example, I don't know if a clean mechanism for picking code atB boot time is available.  And drivers are rather likely to be where$ byte-length instructions are wanted.  I For short instruction sequences, you could code all the versions into the.I module and branch to the "right" one at run time.  In small doses, that'sOJ likely not a problem.  You'd pay for the extra branches on every platform,J put hopefully the price is minimal.  It might be difficult to code this isA a HLL; you'd need to change the code generator's preferred set ofv= instructions in mid-module.  I don't recall a way to do that.o   -- t Robert Deininger rdeininger@mindspring.come   ------------------------------    Date: 22 Mar 2001 01:43:43 +0800, From: Paul Repacholi <prep@prep.synonet.com># Subject: Re: Support of old systemsn- Message-ID: <871yrr9fw0.fsf@prep.synonet.com>   4 rdeininger@mindspring.com (Robert Deininger) writes:  C > It might be difficult to code this is a HLL; you'd need to changeCF > the code generator's preferred set of instructions in mid-module.  I  > don't recall a way to do that.  E BUILTIN is your friend :) And you can define it as an inlined routinea for EV4 and 5 variants.-   -- -< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.j@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------    Date: 22 Mar 2001 01:41:28 +0800, From: Paul Repacholi <prep@prep.synonet.com># Subject: Re: Support of old systemsz- Message-ID: <878zlz9fzr.fsf@prep.synonet.com>q  4 rdeininger@mindspring.com (Robert Deininger) writes:  > > In article <87d7bbb5xu.fsf@prep.synonet.com>, Paul Repacholi  > <prep@prep.synonet.com> wrote: >  > J > > BTW, is not the PAL code meant to implement all new Alpha instructions > > on older machines?   > However... > PALcode traps are slow.yD I know that. Other problem is it seems the load byte unaligned takesF infinite time on a 3000-600. Yes SRM is the latest before anyone asks.    > > But in a device driver, for example, I don't know if a cleanD > mechanism for picking code at boot time is available.  And driversD > are rather likely to be where byte-length instructions are wanted.  6 Oh you please, in a driver I'd want them, no question!   -- b< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.h@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  # Date: Wed, 21 Mar 2001 18:33:55 GMTy2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)# Subject: Re: Support of old systems1. Message-ID: <nu6u6.9$fB6.354@news.cpqcorp.net>   In article <y4wv9j8e20.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>, Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> writes:5 :hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) writes:5 :0F :>   Decade-old computers are very slow and very constrained, and are  :>   more expensive to support.s :-G :While true, some have widgets that aren't available for newer systems,lD :and as long as they manage to get the work done, there is no strongH :reason to replace them. Have a look at the real-time system controllingC :the space shuttle - the design was old when it first flew in 1981.o  G   The AP-101 series (AP-101B, AP-101S) general purpose computers are a gC   descendent of an IBM 360 series (the System/4 Pi, IIRC), and wereiH   specifically designed and specified for the application requirements, G   not the least of which involved the operations of the systems within  K   the target environment -- AP-101B featured ferrite core memory, a design eH   that is tolerant of radiation and that also maintains its memory with B   all power off.   (One of the five shuttle AP-101B GPC series wasG   "frozen", and left powered down but containing the code and the data mE   necessary for return from orbit.  AP-101S lacks this mode, but does]B   have a low-power "sleep" mode.)  Also of interest to the shuttleC   designers were the existing uses of and long-term support of the sH   AP-101 GPCs in various military aircraft including the B-1 and B-52.    E   As part of this GPC selection process, the customer (NASA, in this  J   case) specifically contracted for spare parts and for long-term support.  G   Other customers with embedded or long-term systems requirements also  J   normally take specific steps to ensure spares and long-term maintenance K   -- for a recent example of this, please see the contractual requirements O   involved with DII COE.  A   Another class of customers involve those that find a particulardI   configuration sufficient for requirements, and are particularly loathe EF   to upgrade or otherwise change the configuration.  This is certainlyE   a reasonable approach, but it must be recognized that this approach    does defer the costs.R  K :>   Tru64 UNIX specifically decided to discontinue support of TURBOchanneleL :>   systems, and to focus engineering and qualification efforts on PCI and ( :>   new widget and new feature support. :sL :But the old stuff should continue to work, subject to resource constraints.  D   Why?  At what cost?  Support for older hardware can and does placeF   constraints on new work and new development, in addition to the veryF   obvious detractions involving testing and engineering support costs.H   Case in point: CD-ROM media on VAX.  Without this, we are required to K   maintain a gazillion tape bootstraps and a parallel tape-based bootstrap 1J   environment and a tape-based installation.  Were OpenVMS able to switch B   over to CD-ROM distributions, there could have made a number of L   improvements made to the OpenVMS VAX installation and upgrade environment.E   And OpenVMS VAX would have just one system bootstrap environment...   I :These two reasons are much stronger, IMO, than the others. Are among thedL :elegant ways little macros that allow the boot strap code to patch the restI :of the exec to use the appropriate instruction sequences - like the "bigTI :memory" (more than 32k PFNs) support did on the VAX? (One of the neatestf :hacks I've seen, BTW.)X  J   The instruction sequences and generated codepaths are far different and 5   are far more involved than was the VAX XPA support.      In other words, TANSTAAFL.  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Wed, 21 Mar 2001 12:48:36 +0100a, From: "Bart Zorn" <B.Zorn@TrueBit.n0spam.nl> Subject: Re: sytem logicalsr> Message-ID: <3ab89534$0$2170$1a0eedbc@dreader2.news.xs4all.nl>  > Well, I assume you mean SYS$SYLOGIN, which usually resolves toI SYS$MANAGER:SYLOGIN.COM. However, this file is a highly unlikely place to   find any DEFINE/SYSTEM commands.  L With respect to your suggestion for a brute force search, most installationsH I have seen use an abbreviation of the various DEFINE/SYSTEM commands inD system startup files. Therefore I do not expect you to find a lot of matches.  L About the only answer that I can think of for the original question is: do aL fresh install of OpenVMS on an empty disk, and after you logon for the firstJ time, do a SHOW LOGICAL/OUTPUT=whatever command, before you do any further configuration of the system.K Then you have a listing of all logicals which you could consider universal.HK There is, however, no guarantee whatsoever that no OpenVMS system will have  changed one or more of them.  . Like Hoff already asked, "what are you up to?"  	 Bart Zorn-  = "William Hymen" <t18_pilot@hotmail.spam.com> wrote in message 2 news:2QTt6.80462$lj4.2182571@news6.giganews.com...) > If you have the authorization to do so,t, > search through sylogon for "define/system"= > or just do a brute force search using disk:[000000...]*.coms > define,sys/match=and > Bill >00 > "Tom Linden" <tom@kednos.com> wrote in message5 > news:CIEJLCMNHNNDLLOOGNJIAELECDAA.tom@kednos.com...II > > Sorry if this is an obvious question, but I am looking for a documentt that > > would listL > > the system logicals that are universal to all VMS installations, e.g. is > > SYS$LIBRARY L > > a universal and unambiguous logical in all versions of VMS from say 5.5? > >, > > TIAi > > Tom  > >s >h >t   ------------------------------  % Date: Wed, 21 Mar 2001 06:23:34 -0800h! From: Tom Linden <tom@kednos.com>n Subject: RE: sytem logicalsn9 Message-ID: <CIEJLCMNHNNDLLOOGNJIMEMECDAA.tom@kednos.com>i  ? Thanks for the response, what I was looking to determine was if L SYS$LIBRARY:STARLETSD.TLB existed on all later VMS systems.  Consensus seems: so.  Curiosity caused me to ask the more general question.     > -----Original Message-----3 > From: Bart Zorn [mailto:B.Zorn@TrueBit.n0spam.nl]D) > Sent: Wednesday, March 21, 2001 3:49 AMn > To: Info-VAX@Mvb.Saic.ComI > Subject: Re: sytem logicalsi >n >o@ > Well, I assume you mean SYS$SYLOGIN, which usually resolves toK > SYS$MANAGER:SYLOGIN.COM. However, this file is a highly unlikely place toy" > find any DEFINE/SYSTEM commands. >t@ > With respect to your suggestion for a brute force search, most > installationseJ > I have seen use an abbreviation of the various DEFINE/SYSTEM commands inF > system startup files. Therefore I do not expect you to find a lot of
 > matches. >l< > About the only answer that I can think of for the original > question is: do as@ > fresh install of OpenVMS on an empty disk, and after you logon > for the firstfL > time, do a SHOW LOGICAL/OUTPUT=whatever command, before you do any further > configuration of the system.B > Then you have a listing of all logicals which you could consider > universal.C > There is, however, no guarantee whatsoever that no OpenVMS systemi > will havee > changed one or more of them. >.0 > Like Hoff already asked, "what are you up to?" >s > Bart Zorn- >-? > "William Hymen" <t18_pilot@hotmail.spam.com> wrote in messagee4 > news:2QTt6.80462$lj4.2182571@news6.giganews.com...+ > > If you have the authorization to do so,+. > > search through sylogon for "define/system"? > > or just do a brute force search using disk:[000000...]*.comm > > define,sys/match=and > > Bill > >n2 > > "Tom Linden" <tom@kednos.com> wrote in message7 > > news:CIEJLCMNHNNDLLOOGNJIAELECDAA.tom@kednos.com...cK > > > Sorry if this is an obvious question, but I am looking for a document  > that > > > would list7 > > > the system logicals that are universal to all VMS  > installations, e.g. is > > > SYS$LIBRARYn@ > > > a universal and unambiguous logical in all versions of VMS > from say 5.5?  > > >l	 > > > TIAp	 > > > Tom  > > >e > >w > >d >e   ------------------------------    Date: 21 Mar 2001 08:52:40 -00004 From: Doc.Cypher <Use-Author-Address-Header@[127.1]>A Subject: The art of flaming (was Re: OpenVMS Educational Program)i5 Message-ID: <20010321085240.3382.qmail@nym.alias.net>   " -----BEGIN PGP SIGNED MESSAGE-----  ? On Wed, 21 Mar 2001, Christof Brass <brass@infopuls.com> wrote:e
 >It's sad.  < I know, but you need more practice flaming people Christoff.  K Unfortunately, Google's butchery of the deja archives makes showing you thefJ spanking I gave Tony Szopa a little difficult. Just watch out for mentions< of my character (Lord Running Clam) on the following page...  K http://groups.google.com/groups?num=25&hl=en&lr=&group=sci.crypt&safe=off&tU  h=21e9c8e27f8e6d97&start=75&ic=1  H Yes, that eventually led to M$ being called "The Biggest Humbug you ever
 done saw".  K N.B. I said M$, not Sun or any other Un*X producer. I may not be incrediblytI fond of an OS where they practice command set security through obscurity,n but it beats M$ anyday.n     Doc.   -----BEGIN PGP SIGNATURE-----e Version: 2.6.2  @ iQEVAwUBOrfg8sriC3SGiziTAQFn+wf/RN0xlDFXdLxe5U+eNf3r2rTZNG/OJABz@ Wdw8tjLp41Hy958HAxKV51eGrxh2EcuHHxaTNHBn8iZTrS4J38tHTJ17ECRTx7Jt@ RC2GXhvSAdPReMnFBBCvtTRtk8sfaIWqTc4vJ9nju1u0Z1gJfELjZSTwFpWj7FF/@ CLX+QxEE0UaYpEWs8Fkg5Z1DWtuSF1aPm0/w+c8XM5I/q/vWr+MSgLxfGslZ+FUn@ wLnVdKHeiw+Vv4u2wymWasUODMMziT9vsjfSfcgMQMfncW8nkT2Vz2e0174wHc2A8 zXetIXPpv94HJTH+PI01twYDutXF0Gw3bYyVAo8HqpO2WRr5zbFI1g== =OobL  -----END PGP SIGNATURE-----s   ------------------------------    Date: 21 Mar 2001 14:21:39 +0100G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>s" Subject: Re: TURBOchannel graphicsH Message-ID: <y4u24n8dgc.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  - Arne Bergseth <Arne.Bergseth@dnv.com> writes:o  I > > >ftp://ftp.netbsd.org/pub/NetBSD/misc/dec-docs/index.html#EK-369AA-OD-? > The URL abowe was pasted from the address line of my browser.r5 > It works for me with Netscape 4.7 as it is written.:F > The reference is to a HTML document stored on a FTP server I believe  K Yes, but the ...#... pointing to a particular label inside the HTML file ispK not particularly useful to FTP...another of these "the user's stupid, let'spK guess what he means and do that" features (like prepending http:// and thene. inserting www if you don't write it yourself).   	Jan   ------------------------------  % Date: Wed, 21 Mar 2001 13:34:33 -0000t* From: "Richard Brodie" <R.Brodie@rl.ac.uk>" Subject: Re: TURBOchannel graphics, Message-ID: <99aald$1eio@newton.cc.rl.ac.uk>  T "Jan Vorbrueggen" <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> wrote in messageB news:y4u24n8dgc.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de.../ > Arne Bergseth <Arne.Bergseth@dnv.com> writes:t  H > > The reference is to a HTML document stored on a FTP server I believe >-M > Yes, but the ...#... pointing to a particular label inside the HTML file iscM > not particularly useful to FTP...another of these "the user's stupid, let'siM > guess what he means and do that" features (like prepending http:// and thena0 > inserting www if you don't write it yourself).  J No, it's the result of specifying the syntax of URI's independently of theH protocol. I don't see the problem: if you're going to use something thatB understands the URI format, it should understand fragment IDs. The" transfer protocol is not relevant.   ------------------------------  % Date: Wed, 21 Mar 2001 18:52:18 +0100   From: Paul Sture <paul@sture.ch>" Subject: Re: uVax3100 m10 & VMS5.2+ Message-ID: <VA.00000320.58aaaf96@sture.ch>0  D In article <j8dcbt0qqvf7vj1466r4mtr471722uie9k@4ax.com>, Greg elkin  wrote:5 > From: greg elkin <cmkrnl@bouncy-castle.demon.co.uk>z > Newsgroups: comp.os.vmst$ > Subject: Re: uVax3100 m10 & VMS5.2' > Date: Mon, 19 Mar 2001 16:35:08 +0000T > & > Ah, there they are, on the web site!* > http://www.montagar.com/htbin/ohp_hobreg >  > :-)h > F Just don't forget to do it twice - once for the VMS license, and then  again for the Layered licenses.q   ___f
 Paul Sture Switzerland.   ------------------------------  % Date: Wed, 21 Mar 2001 12:56:14 -05001- From: "Peter Weaver" <peter.weaver@stelco.ca>rY Subject: VERB SHOW (was Re: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS Educati14 Message-ID: <rX5u6.163329$Z2.2017043@nnrp1.uunet.ca>  4 Trying to avoid this thread as much as possible but;  > "Bill Gunshannon" <bill@triangle.cs.uofs.edu> wrote in message& news:99aj5s$2rjg$1@info.cs.uofs.edu...8 > In article <1010321054351.21646A-100000@Ives.egh.com>,% >  John Santos <JOHN@egh.com> writes:u >...? > |> a program to execute.  Unless all the "show" commands were  executed byO > |> a single program, etc.) >o3 > You mean like they are on VMS??  [SYSEXE]SHOW.EXEr  : SHOW INTRUSION uses the image CIA, SHOW ACL uses the imageB SETSHOSECUR, SHOW CPU uses the image SMPUTIL. There are other SHOWF commands that invoke other images. None of these commands use the SHOW image.   >s >...F > First, the commands (ie. ls, ps, rm, ln, mv, etc.) are not a part ofA > the shell int he first place any more than SHOW is part of DCL.u  ? SHOW DEFAULT is a CLIROUTINE, the same as SHOW PROTECTION, SHOWtD SYMBOL, SHOW TRANSLATION and others. These commands do not invoke anA image. That is why you can control-Y out of an image, issue thesel= commands and then CONTINUE your image again. I learnt that ontC 13-NOV-1984 on VMS 3.7 but the story is much too long to post here.n   ------------------------------    Date: 22 Mar 2001 01:29:04 +0800, From: Paul Repacholi <prep@prep.synonet.com>5 Subject: [Announce] Perth WA LUG reformation meeting.e- Message-ID: <87d7bb9gkf.fsf@prep.synonet.com>r  ? A meeting for all those interested in re-starting the Perth LUGmB will be held on Tues 10-Apr, at 5:30 for 6. Venue is the Blue Note Taven in Colin St. West Perth.  G Please pass this information on as far as possible, and hope to see yout there.  ) RSVPs or queries to prep@prep.synonet.como   -- -< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.:@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  % Date: Wed, 21 Mar 2001 10:15:49 +0000n From: Roy Omond <Roy@Omond.net>t? Subject: [Change subject] What if VMS and MacOS had prevailed ?a) Message-ID: <3AB87F55.706FACFB@Omond.net>e   [... everything snipped ...]   <Soapbox-mode On>o  = It's always caused me some sadness to observe the tendency to < get into bed with Micros**t and the PC world, rather than to< pursue the (natural) affinity between VMS and the Macintosh.  8 Just think what might have been achieved if the original= enthusiasm between the 2 quality environments (VMS and MacOS)r= had prevailed (remember the days of Pathworks for Macintosh),o? rather than the (IMHO) the painful attempt to integrate quality.= (VMS) with the, ahem, "sub-standard" ever-moving goalposts ofn
 the PC world.   C The IT world is littered more than any other with "wrong decisions"e' which will bite us time and time again.T   <Soapbox-mode Off>   Ah well ...m  	 Roy Omondl Blue Bubble Ltd.   ------------------------------   Date: 21 Mar 2001 07:10:40 GMT- From: djweath@attglobal.net (Dave Weatherall) $ Subject: Re: [DCL] minute of the day5 Message-ID: <DTiotGxQ0bj6-pn2-KSDdn7yTId3u@localhost>   B On Tue, 20 Mar 2001 15:47:21, Jim Agnew <agnew@hsc.vcu.edu> wrote:   > I remember on a pdp-11, losing a whole years worth of FORTRAN code...  Right in front of my micromanaging women's libber boss... . > (Now, there are people who are reasonable, and those who are UNreasonable..)  I acted normal, went home, then sneaked back and^ > edited all the compiler listings into source...  whew... and they never found out, either... > . > If it happens to you once, you REMEMBER!!!!! >  > Jim. >    I wonder if she reads c.o.v :-)-   --   Cheers - Dave.   ------------------------------   Date: 21 Mar 2001 02:32 CSTo' From: carl@gerg.tamu.edu (Carl Perkins)m$ Subject: Re: [DCL] minute of the day- Message-ID: <21MAR200102321414@gerg.tamu.edu>h  0 Didier Morandi <Didier.Morandi@gmx.ch> writes... [...]a= }$ if p1 .eqs. "" then inq p1 "_File(s) (wild cards allowed)"n }$ if p1 .eqs. "" then exitH }$ file = p1 }$ say = "write sys$output"t% }$ if f$locate(".",p1) .eq. f$len(p1)  }$ then  }$    say "Invalid syntax: ",p1 
 }$    exit }$ endif }$ name = f$element(0,".",p1)b1 }$ if f$locate("*",name) .eq. f$len(name) .and. - A }     f$locate("%",name) .eq. f$len(name)  then goto NO_WILDCARDSH- }$ files_list = "sys$scratch:files_list.temp" Q }$ dire_/col=1/notrail/nohead/out='files_list' 'p1'.     !the "." is mandatory!!!1 }$ close/nolog chP [...]g  A All this do-it-yourself filename parsing is not a very good plan.t  C You'd get better results using F$Parse, and also instead of reading < directory output back in it would be better to use F$Search.  B For example, what happens in the above if, for P1, the user enters> [over.there]*.out? You end up looking for files named "[over".  9 Using the supplied lexical functions is much more robust.t i.e. $ name = F$Parse(P1,,,"NAME")vE will find the file name regardless of details if it is passed a valid  file specification.n  	 Examples:   5 $ say f$parse("disk:[over.there]*.out",,,"directory")d [OVER.THERE]0 $ say f$parse("disk:[over.there]*.out",,,"name") *o0 $ say f$parse("disk:[over.there]*.out",,,"type") OUT    --- Carl   ------------------------------  % Date: Wed, 21 Mar 2001 09:06:50 -05001# From: Jim Agnew <agnew@hsc.vcu.edu> $ Subject: Re: [DCL] minute of the day+ Message-ID: <3AB8B57A.94BEAAAA@hsc.vcu.edu>o  = ain't seen her in years...  ;-)  i'm assuming it's safe now..h   ------------------------------  % Date: Wed, 21 Mar 2001 18:52:18 +0100c  From: Paul Sture <paul@sture.ch>$ Subject: Re: [DCL] minute of the day+ Message-ID: <VA.00000321.58aab0b8@sture.ch>u  < In article <3AB77B89.3594AFBA@hsc.vcu.edu>, Jim Agnew wrote:% > From: Jim Agnew <agnew@hsc.vcu.edu>a > Newsgroups: comp.os.vmse& > Subject: Re: [DCL] minute of the day' > Date: Tue, 20 Mar 2001 10:47:21 -0500n2 > To: Robert Deininger <rdeininger@mindspring.com> >  > I remember on a pdp-11, losing a whole years worth of FORTRAN code...  Right in front of my micromanaging women's libber boss... i > (Now, there are people who are reasonable, and those who are UNreasonable..)  I acted normal, went home, then sneaked back and^ > edited all the compiler listings into source...  whew... and they never found out, either... > . > If it happens to you once, you REMEMBER!!!!! > L Back in PDP days, we lost most of week's development once, the morning of anP Open Day when I was supposed to give a demo of our latest shiny new application.  F We plied our audience with specially imported _strong_ beer before theM presentation, and I waffled quite extensively around login screens and menus,pP hiding the fact that it was mostly broken behind the scenes. We got away with it :-)t  M Paper listings came to our aid, except for the junior programmer who had justoL finished his first large program, and thrown his listings away to celebrate,G thinking he'd grab new ones on the Monday. He was _not_ a happy chap...e ___ 
 Paul Sture Switzerlandt   ------------------------------    Date: 21 Mar 2001 12:22:17 +0100G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>t3 Subject: Re: [Q] TCP/IP RPC connection error status2H Message-ID: <y4elvr9xjq.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  . Didier Morandi <Didier.Morandi@gmx.ch> writes:  . >  transp = svctcp_create(RPC_ANYSOCK, 0, 0) ; >         if (transp == NULL)p > [...] I wish to do this:2 >      transp = svctcp_create(RPC_ANYSOCK, 0, 0) ;& >      if((transp & 1) != SS$_NORMAL) ; >         fprintf(stderr, "cannot create tcp service.\n") ;n >         lib$stop(transp);   M If the initial code is correct, svctcp_create returns a pointer, not a statuseM value, so your substitute code won't work as intended. Follow the errno etc. - route offered by Bob.a   	Jan   ------------------------------  % Date: Wed, 21 Mar 2001 12:01:07 -0500 * From: "Andy Stoffel" <acs@fcgnetworks.net>L Subject: [Random non-DCL VMS quesion of the day] Need a clue about CLUE help8 Message-ID: <z75u6.82683$lj4.2249624@news6.giganews.com>  1 Just wondering why in VMS 7.2-1 in ANALYZE/SYSTEM   : CLUE HELP and HELP CLUE don't list the exact same options. This seems a bit "jarring"...h  
 For example :o   $ ANALYZE/SYSTEM" OpenVMS (TM) Alpha system analyzer   SDA> HELP CLUE   CLUE  #   Additional information available:p  K   CALL_FRAME CLEANUP    CONFIG     CRASH      ERRLOG     FRU        HISTORY J   MCHK       MEMORY     PROCESS    REGISTER   SG         STACK      SYSTEM   VCC        XQP   CLUE Subtopic?  	 -- and --S   SDA> CLUE HELP   CLUE    G   CLUE (Crash Log Utility Extractor) facilitates crash dump analysis byhK   automatically collecting and archiving commonly needed information from a G   dumpfile. CLUE is called through SDA extensions from the standard SDAe prompt:m  4         SDA> CLUE <clue-command> [/<clue-qualifier>]      #   Additional information available:8  K   Logicals   CLEANUP    CONFIG     CRASH      DEBUG      ERRLOG     HISTORYgG   KPB        MCHK       MEMORY     PROCESS    SCSI       STACK      VCCu   XQPg   CLUE Subtopic?   --= Based on a time difference (and disk activity I can here when = I do CLUE HELP)I'm assuming that one ( HELP CLUE ) is gettingcC it's help text from SDA's help file & CLUE has it's own help file ?c  " Correct ? Or am I just CLUE-less ?   Thanks   -Andy-   ------------------------------  % Date: Wed, 21 Mar 2001 17:51:28 +0000t  From: steven.reece@quintiles.comP Subject: Re: [Random non-DCL VMS quesion of the day] Need a clue about CLUE helpH Message-ID: <OF5A735F15.F2ADC341-ON80256A16.0061FE64@qedi.quintiles.com>  H Having looked at both libraries using HELP/LIBRARY= I would suggest that7 you're right Andy.  Now why is a different matter......l Steve.   Andy Stoffel wrote:O@ >>>Based on a time difference (and disk activity I can here when= I do CLUE HELP)I'm assuming that one ( HELP CLUE ) is gettingsC it's help text from SDA's help file & CLUE has it's own help file ?   % Correct ? Or am I just CLUE-less ?<<<    ------------------------------   End of INFO-VAX 2001.160 ************************